BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//TYPO3/NONSGML Calendar Base (cal) V1.4.2//EN
METHOD:PUBLISH
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_238
DTSTAMP:20100204T102024
DTSTART:20100204T140000Z
DTEND:20100204T150000Z
CATEGORIES:DepKo
SUMMARY:DepKo: Multifractal processes to model images and texture
DESCRIPTION:Multifractal processes to model images and texture\n\nPierre ChainaisDuring the last ten years, various works have provided a variety of stochastic processes with nice properties such as scale invariance and non Gaussian statistics. Such processes belong to the family of multifractal processes...They have been generalized to N dimensions, which gives rise to the possibility of image (2D) or porous media (3D) modeling for instance. Moreover, these processes can be numerically synthesized which results very useful for applications. We will present the family of multifractal processes and focus on the construction of infinitely divisible cascades in particular. We will show that they share most of the statistical properties of natural images and present an application to the modeling of images of the Sun taken by the spatial telescope EIT. We will also evoke an application to texture synthesis. Host: Matthias BethgeContact:Matthias BethgePhone: +49(0)7071-601 1770Email: mbethge@tuebingen.mpg.de\n\nFor more information please check the Google Calendar...\n\n
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_246
DTSTAMP:20100226T163653
DTSTART:20100316T101500Z
DTEND:20100316T110000Z
CATEGORIES:DepKo
SUMMARY:Depko: Eight illusions of color and motion 
DESCRIPTION:Prof. Stuart Anstis, University of California, San DiegoSome new(-ish) illusions of visual colour and motion give clues as to how we normally process colour and motion.Colour: Afterimages are usually complementary to the adapting colours. But I shall show that one and the same inducing stimulus can produce afterimages of different colours. Afterimages are enhanced, and their colours potentially changed or averaged within contours, when they are projected on a white test field that contains black line contours. I shall also demonstrate coloured afterimages generated by induced colours, and even by unseen colours. Motion: In the Footsteps illusion, high contrast objects seem to move faster than low contrast objects. In the Zigzag illusion, size matters: a pattern that appears to move to the right when viewed from a distance appears to move downwards when viewed from close up. In the Chopsticks and Rings illusions, intersecting contours can be seen as either rigid or sliding and their perceived directions of movement and our ability to track these movements with our eyes, vary accordingly. These illusions show the interaction between low-level signals from neural motion detectors, and top-down signals aimed at parsing moving objects.  Contact:Dr. Marc ErnstPhone: +49 7071 601-644E-Mail: marc.ernst@tuebingen.mpg.de
LOCATION:Max Planck House Lecture Hall
ORGANIZER;CN="Marc Ernst":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_247
DTSTAMP:20100226T164252
DTSTART:20100316T130000Z
DTEND:20100316T140000Z
CATEGORIES:DepKo
SUMMARY:Depko: The contribution of the motor system to the understanding of action and language
DESCRIPTION:Dr. Luciano Fadiga, University of Ferrar, Italian Institute of Technology, ItalyThe contribution of the motor system to the understanding of action and languageIn the recent years we have empirically shown that the observation of actions performed by others automatically activates the observer's premotor cortex. These data, coming from electrophysiological recordings in monkeys (mirror neurons) and from TMS and brain imaging studies on humans, support the idea that this observation-evoked motor resonance could be the functional expression of a brain mechanism involved in understanding the actions made by others. However, it might appear paradoxical that the motor centres, classically considered as the output stage of the brain, are involved in such high-level cognitive task. In the first part of my presentation I will discuss this issue by showing that mirror neurons are indeed part of a more general mechanism involving motor representations in perceptual processing. In the second part of my talk I will show that during speech listening TMS reveals an enhancement of excitability of the listener's tongue motor representations and, according to some very new data, this motor involvement contributes to speech perception. Finally, I will discuss the involvement of Broca's area in action understanding, being this speech-related centre constantly activated by the observation of actions of others. More Information...
LOCATION:Max Planck House Lecture Hall
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_275
DTSTAMP:20100329T092621
DTSTART:20100402T080000Z
DTEND:20100402T090000Z
CATEGORIES:DepKo
SUMMARY:Friko: llumination Perception in 3D Scenes - Holly Gerhard (New York University, USA)
DESCRIPTION:llumination Perception in 3D Scenes - Holly Gerhard (New York University, USA)\n\nThe light signal falling on the retina confounds illumination and object properties, causing the fundamental problem of surface property estimation to be under-constrained. In  everyday 3D environments where illumination is inhomogeneous, the light signals reflected from an object with constant surface properties can vary greatly depending on local surface orientation and location within the environment. Psychophysical research in the past ten years has demonstrated that human observers infer the spatial variation of the light field when judging object properties and that they can make reliable judgements about the light field directly. In the first part of the talk, I will present results demonstrating that the human visual system represents light field variation with good temporal and spatial resolution. We presented human observers with rendered scenes of matte 3D objects illuminated by  dynamic out-of-view light sources and viewed stereoscopically. The only cue to illuminant spatial variation was the surface shading on the objects, from which observers were required to infer light field dynamics. Observers were sensitive to information about the light field not captured by local image statistics but that required global processing of the scene’s 3D structure, and this sensitivity was found to enhance detection of surface property changes. In the second part of the talk, I will present an analysis of how human observers estimate illumination direction from shape and shading, based on the errors made in a similar lighting motion discrimination experiment.\n\nHost: Matthias Bethge\n\n
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_249
DTSTAMP:20100226T164704
DTSTART:20100413T091500Z
DTEND:20100413T100000Z
CATEGORIES:DepKo
SUMMARY:Depko: Computational Principles of Adaptive Motor Control
DESCRIPTION:Dr. Daniel Braun, University of Oxford, United KingdomComputational Principles of Adaptive Motor Control AdaptabilityComputational Principles of Adaptive Motor Control Adaptability is an essential feature of human motor control in the face of complex and variable environments. Conceptually, such adaptive control can be broken down into structural and parametric learning. Structural learning extracts invariants of sensorimotor mappings that could be used, for example, to guide exploration when a novel, but similar control task has to be learned. We will discuss evidence that structure learning might provide a basis for learning-to-learn and skill generalization. Adaptive optimal feedback control will be discussed as a framework for parametric adaptive control. We also briefly mention learning in motor game tasks where the environment is given by another actor. In the second part of the presentation we will discuss a minimum relative entropy principle as a potential normative basis for adaptive control. This leads to a stochastic Bayesian control rule where controls are sampled from a posterior distribution over a set of probabilistic input-output models and the true model is identified by Bayesian inference. This turns the adaptive control problem into inference and sampling processes that work both forward in time and hence are applicable on-line. We will discuss simple applications. More about Dr. Daniel Braun...Contact:Dr. Daniel BraunPhone: +44 1223 7 48512E-mail: dab54@cam.ac.uk
LOCATION:Max Planck House Lecture Hall
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_250
DTSTAMP:20100226T164912
DTSTART:20100421T130000Z
DTEND:20100421T140000Z
CATEGORIES:DepKo
SUMMARY:Depko: Signals and 'noise' in olfactory circuits
DESCRIPTION:"Signals and 'noise' in olfactory circuits"Prof. Gero Miesenboeck; Department of Physiology, Anatomy and Genetics, Magdalen College, Oxford UK\n\nCircuits of the antennal lobe, the insect equivalent of the olfactory bulb, translate input from olfactory receptor neurons (ORNs) into projection neuron (PN) output. Analyzing Drosophila mutants lacking ORN input to one glomerulus, we and oth-ers have shown that PNs innervating a particular glomerulus remain responsive to odors even when all direct ORN afferents to that glomerulus are silenced. PNs thus receive excitatory input from neurons other than ORNs. The likely source of this additional input is a previously unidentified population of excitatory local neu-rons (eLNs) that densely innervate the antennal lobes. These eLNs respond broadly to odors but exhibit little glomerular specificity in their synaptic output, suggesting that the receptive properties of PNs reflect a combination of glomeru-lus-specific ORN afferents and diffuse LN excitation. The discovery of eLNs raises a puzzling question: Why add a diffuse offset to make PN firing less odor-specific than that of its ORN inputs? One possibility is that the excitatory back-ground generated by eLNs serves to boost faint odor-specific signals, in a mechanism reminiscent of stochastic resonance. This hypothesis predicts that silencing the eLN amplifier raises the thresholds for both Kenyon cell activation and olfactory-driven behaviors. The hypothesis demands as a corollary that eLNs are active at low odor concentrations, when the need for signal amplification is largest. I present experimental evidence, obtained using the optogentic approaches pioneered by my group, that supports all of these predictions.\n\n
LOCATION:Max Planck House Lecture Hall
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_262
DTSTAMP:20100305T131322
DTSTART:20100428T091500Z
DTEND:20100428T100000Z
CATEGORIES:DepKo, FriKo
SUMMARY:Depko: The coupling between action and attention
DESCRIPTION:Prof. Dr. Glyn Humphreys - School of Psychology, University of Birmingham, United Kingdom The coupling between action and attention  Though it has often been argued that action and perception can dissociate, there is increasing evidence that attention and perception are modulated by action. Here I will report results demonstrating a close coupling between action-related information in the environment and attention, based on effects when (i) two objects appear to interact together, and (ii) when objects are presented with hands that grasp or do not grasp the objects correctly. Drawing on evidence from neuropsychology, experimental psychology and functional brain imaging, I will argue that these action-related effects involve differential visual as well as motor responses to stimuli, indicating that our visual systems are tuned to action-related properties of the environment.Host: Lewis Chuang
LOCATION:Max Planck House Lecture Hall
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_287
DTSTAMP:20100429T113116
DTSTART:20100504T091500Z
DTEND:20100504T100000Z
CATEGORIES:DepKo
SUMMARY:Depko: Modeling Prosopagnosia: Computational Theory and Experimental Investigations of a Deficit in Face Recognition
DESCRIPTION:Dr. Rainer Stollhof - Max-Planck-Institut für Mathematik in den Naturwissenschaften\n\n\n\nModeling Prosopagnosia: Computational Theory and Experimental Investigations of a Deficit in Face Recognition\n\n\n\nProsopagnosia is a profound deficit in facial identification which can\n\nbe either acquired due to brain damage or is present from birth, i.e.\n\ncongenital. Normally, faces and objects are processed by different\n\ncortical systems, and the spatial separation underlies the selectivity\n\nof the deficit in acquired prosopagnosia. This separation is still\n\npresent in congenital prosopagnosia (CP) and questions modular accounts of the deficit. \n\n\n\nIn this talk, I will present a computational model of CP that was\n\nderived from formal considerations, implemented in artificial neural\n\nnetwork models of facial information encoding, and tested in experiments\n\nwith CP subjects. The main hypothesis is that the deficit in CP is\n\ncaused by a failure to obtain adequate descriptions of individual faces:\n\nA predisposition towards a reduced structural connectivity in visual\n\ncortical areas enforces descriptions of visual stimuli that lack the\n\namount of detail necessary to distinguish a specific exemplar from its\n\npopulation, i.e. achieve a successful identification. \n\nIn contrast to previous models of acquired prosopagnosia, here, the\n\ndevelopmental aspect of CP is incorporated explicitly into the model,\n\nquantitative arguments for a deficit that is task specific\n\n(identification) - and not necessarily domain specific (faces) - are\n\nprovided for synthetic as well as real data (face images), and the model\n\nis validated empirically in experiments with CP subjects. \n\nFinally, I will provide a summary of an extensive and detailed empricial\n\ninvestigation of face and object recognition deficits in CP that enables\n\na better behavioral characterization and the identification of deficit-specific subtypes of CP.\n\nHost: Isabelle Bülthoff\n\n ~~~~~\n\n\n\nMax-Planck-Institut für Mathematik in den NaturwissenschaftenInselstr. 22\n\n04103 Leipzig Germany \n\n\n\n\n\n
LOCATION:Max Planck Haus Lecture Hall
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_251
DTSTAMP:20100226T165146
DTSTART:20100510T091500Z
DTEND:20100510T100000Z
CATEGORIES:DepKo
SUMMARY:Depko:Multimodal Probes for Molecular Imaging
DESCRIPTION:Multimodal Probes for Molecular ImagingAngelique Louie, Associate Professor, Department of Biomedical Engineering, University of California, Davis\n\n\n\nAbstract:\n\nThe conundrum of modality selection in clinical diagnostic imaging is that modalities with the highest sensitivity have relatively poor resolution, while those with high resolution have relatively poor sensitivity. In recent years, the idea of using multiple modalities in conjunction has gained in popularity and researchers have come to realize that the complementary abilities of different imaging modalities could be harnessed to great effect by using them in tandem. The idea of combining imaging technologies moved to the mainstream with the advent of the first successful commercial fused instruments. The first fused PET/CT instrument, developed in 1998 by Townsend and colleagues in collaboration with Siemens Medical, was available commercially in 2001. Over the ensuing years PET/CT sales increased with such vigor that by the year 2006 there were virtually no sales of standalone PET instruments; all PET sales were as part of multimodality systems. Now PET/MRI instruments are poised to enter the clinic. With hybrid technology clearly on the rise, the excitement over these new instruments has triggered a tumult of activity in probe design and development in an effort to boost the clinical benefits of hybrid instrument technology. In this talk I present an overview of approaches to develop multimodal probes using current projects in my laboratory as examples. Activatable probes, multimodal quantum dots, and PET/MR imaging of inflamed plaques are discussed.\n\n\n\nLink: http://louielab.bme.ucdavis.edu/~louielab/\n\n\n\nContact:\n\nAngelique Louie\n\nAssociate Professor\n\nDepartment of Biomedical Engineering\n\nUniversity of California\n\nDavis, CA 95616\n\n(530)752-7134 office\n\n(530)754-5739 fax\n\naylouie@ucdavis.edu \n\n\n\nHost: Goran Angelovski
LOCATION:Max Planck House Lecture Hall
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_253
DTSTAMP:20100226T165515
DTSTART:20100517T121500Z
DTEND:20100517T130000Z
CATEGORIES:DepKo
SUMMARY:Depko: The cost of time in motor control
DESCRIPTION:Dr. Reza Shadmehr, Johns Hopkins University, USA\n\nThe cost of time in motor control\n\nThere is consistency in how health people move their eyes, arms, and legs.  What is good about this way of moving, and why has our brain settled on this pattern?  Here, I focus on the control of eye movements and suggest that the purpose of any movement is to acquire a more rewarding state.  I suggest that the way the brain discounts reward in time explains why we move the way that we do.  This temporal discounting framework has the potential to explain why disorders that affect processing of reward in the brain, like Parkinson's disease, depression, and Schizophrenia, result in changes in control of eye movements.\n\n\n\nHost: Marc Ernst
LOCATION:Max Planck House Lecture Hall
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_252
DTSTAMP:20100226T165316
DTSTART:20100518T081500Z
DTEND:20100518T090000Z
CATEGORIES:DepKo
SUMMARY:Depko: Computational stages and cortical connectivity during language production revealed with intra-cranial electrophysiology
DESCRIPTION:Language is uniquely human,  so neuroscientists have so far been\n\nunable to study its basis in the brain with the high-resolution (but\n\ninvasive) methods that are possible in animals. This is one of the\n\nreasons we lack well-grounded neural network models of speech\n\nproduction, and why regions like Broca' area have been associated with\n\nso many different aspects of language.  Using intra-cranial\n\nelectrophysiology (ICE) in pre-surgical patients, I recorded separate\n\nlanguage processing stages for the meaning, structure and sound forms\n\nof words. These were separable in time (peaks at 200, 320, 450ms) and\n\npartly overlapping in space within Broca's area (Science, 2009).  The\n\nspatiotemporally distinct circuits, however, fell within the volume of\n\na typical fMRI voxel and would not have been distinguishable with\n\nconventional neuroimaging. n another group of patients, I found that\n\noscillating brain circuits at a given time and location process\n\ndifferent information at different spectral frequencies, and could\n\nidentify a flow of information from early visual input (~60ms) to\n\narticulatory output (~600ms and beyond) with multiple waves of\n\nsystemwide activity suggesting distinct computations. Overall, the\n\nfindings give a glimpse into a complex level of brain organization in\n\ntime, space, and physiology, which is not visible with fMRI or MEG and\n\nmust be probed much further to understand how the human brain computes\n\nlanguage.
LOCATION:Max Planck House Lecture Hall
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_297
DTSTAMP:20100606T074921
DTSTART:20100608T091500Z
DTEND:20100608T100000Z
CATEGORIES:DepKo
SUMMARY:Depko: Do humans have mirror neurons?
DESCRIPTION:\n\n\n\nDr. James Kilner - The Wellcome Trust Centre for Neuroimaging, UK\n\n\n\nIs it possible to understand the intentions of another person by simply observing their movements? Many neuroscientists believe that this ability is made possible by the brain's mirror-neuron system,through its direct link between action and observation. However, precisely how intentions can be inferred through movement-observation has provoked much debate. In particular, in order to infer intentions from movements we have to be able to solve an ill-posed inverse problem: the problem that the same movement can be caused by many different intentions. In this talk I will provide evidence from behavioural, fMRI and EEG/MEG studies that the function of mirror-neuron system is best considered within a predictive coding framework that appeals to the statistical approach known as empirical Bayesian inference. Within this scheme the most likely cause of an observed action can be inferred by minimizing the prediction error at all levels of the cortical hierarchy that are engaged during action observation. This account identifies a precise role for the mirror-neuron system in our ability to infer intentions from actions and provides the outline of the underlying computational mechanisms.  -- Dr. James Kilner Wellcome Trust Research Career Development Fellow, The Wellcome Trust Centre for Neuroimaging Institute of Neurology 12 Queen Square London WC1N 3BG Great Britain
LOCATION:Max Planck House Lecture Hall
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_300
DTSTAMP:20100614T071551
DTSTART:20100614T090000Z
DTEND:20100614T100000Z
CATEGORIES:DepKo
SUMMARY:Passivity-based control for multiple nonholonomic mechanical systems and multiuser haptics over the Intenet
DESCRIPTION:Dongjun Lee, Dept. of Mechanical, Aerospace & Biomedical Engineering, University of Tennessee - Knoxville, USA\n\n \n\nAbstract:\n\nThe first part of this talk will be on our recently-proposed nonholonomic passive decomposition (NPD), which enables us to decompose Lagrange-D'Alembert dynamics of multiple nonholonomic mechanical systems into: 1) shape system describing formation aspect among the agents; 2) locked system maneuver aspect of their bulk motion as a whole; and 3) quotient system simultaneously perturbing these two aspects.  NPD also preserves Lagrangian dynamics structure and passivity, thereby, greatly facilitating control design/analysis for the decomposed systems.  \n\nWe will introduce geometric and passivity properties of NPD, along with notions of decoupled controllability (i.e. if formation or maneuver can be driven without perturbing the other?) and simple passivity-based formation/maneuver control design example.  Semi-autonomous tele-driving of multiple wheeled mobile robots using NPD and passive set-position modulation (PSPM) framework will also be discussed.\n\n \n\nWe will then briefly cover our recent peer-to-peer distributed control architecture for multiuser shared haptic collaboration in the cyberspace, which synchronizes each user's local copy of the shared virtual environment to provide consistent haptic experience among the users while enforcing passivity for robust-stability/scalability/portability among heterogeneous devices/users.  \n\nNetwork optimization issue will also be discussed with experimental results.\n\n \n\nBio:\n\nDongjun Lee received the B.S. degree in mechanical engineering from Korea Advanced Institute of Science and Technology (KAIST), Taejon, Korea, in 1995, the M.S. degree in automation and design from KAIST, Seoul, Korea, in 1997, and the Ph.D. degree in mechanical engineering from the University of Minnesota at Twin Cities, in 2004.\n\n \n\nSince 2006, he has been an Assistant Professor with the Department of Mechanical, Aerospace and Biomedical Engineering at the University of Tennessee - Knoxville.  He was a Postdoctoral Research Fellow with the Coordinated Science Laboratory at the University of Illinois at Urbana-Champaign from 2004 to 2006, and an Engine Development Engineer at Kia Motors Corporation, Korea, from 1997 to 1999.  His main research interests are dynamics and control of robotic and mechatronic systems with particular emphasis on teleoperation, haptics, multirobot systems, nonholonomic systems and geometric mechanics control theory.\n\n \n\nDr. Lee received the National Science Foundation CAREER Award in 2009, and was a 2002-2003 Doctoral Dissertation Fellow of the University of Minnesota.
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_298
DTSTAMP:20100606T075207
DTSTART:20100616T121500Z
DTEND:20100616T130000Z
CATEGORIES:DepKo
SUMMARY:Depko: TBA - Johannes Burge
DESCRIPTION:Johannes Burge - University of Texas, Austin
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_303
DTSTAMP:20100622T150811
DTSTART:20100623T120000Z
DTEND:20100623T130000Z
CATEGORIES:DepKo
SUMMARY:Depko: Being there without being there: the Challenge of Telerobotics
DESCRIPTION:Being there without being there: the Challenge of Telerobotics\n\n\n\nAbstract: Telerobotics is one of the most intriguing fields of robotics. It allows to break down the barriers of scale and distance, enabling human beings to manipulate objects situated in remote locations and/or to interact with micro/nano-scale environments, otherwise inaccessible. Research in telerobotics has been very active in the last decades and huge theoretical and technological results have been obtained. In this talk the main issues related to telerobotics will be discussed. In particular, it will be shown how energy based control strategies are suitable for implementing bilateral teleoperation systems and for taking into account perceptual issues.\n\n\n\nBio:  Cristian Secchi graduated in Computer Science Engineering, with\n\nmajors in Automation and Robotics, at the University of Bologna in 2000 and in 2004 he earned a Ph.D. in Information Engineering, Curriculum Automation, at the University of Modena and Reggio Emilia where he is currently Assistant Professor. In 2006 he has been selected as one of the three finalists for the Georges Giralt Award for the best PhD thesis on robotics in Europe. He participated to the CROW (Coordination of AGVs in Automatic Warehouses) project that was selected as one of the finalists for the 2010 EUROP/EURON Technology Transfer Award. He has been Associate Editor of the IEEE Robotics and Automation Magazine from 2006 to 2008 and he is co-chair of the IEEE-RAS TC on Telerobotics since 2007. He is the coordinator of the TRAFCON (Traffic Control of AGVs in Automatic Warehouses), Experiment of ECHORD, EU ICT Integrated Project (2010-2011). He is co-author of the book “Control of Interactive Robotic Interfaces: a port-Hamiltonian approach” published by Springer in 2007 and of more than 70 publications about robotics and automation in peer reviewed international journals and conferences.
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_305
DTSTAMP:20100625T122002
DTSTART:20100628T100000Z
DTEND:20100628T110000Z
CATEGORIES:DepKo
SUMMARY:[Depko]: Embodied perspectives in language comprehension 
DESCRIPTION:P.D. Dr. Lars Konieczny, University  of Freiburg\n\n\n\nAbstract:\n\nAccording to the embodied cognition approach to language, comprehension amounts to perceptual and motoric simulation of the described situation (Zwaan 2007, Kaup 2009, Glenberg & Kaschak 2002, Barsallou 1999). I will present behavioral data indicating that comprehenders can take different, and even multiple spatial perspectives in the simulated action scenario depending the indexicality of the thematic roles. Moreover, I will present evidence suggesting that anaphora resolution might be grounded in internalized deictic gestures to discourse objects located in a mental virtual space. Comprehenders interpret nominal expressions by constructing a mental space and canonically locating the referents of the subject and object NP in this imaginary space, where first mentioned (or subject) referents tend to be placed to the left of later referents.\n\n\n\n Host: Tobi Meilinger (AGBU)
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_302
DTSTAMP:20100614T093558
DTSTART:20100630T091500Z
DTEND:20100630T100000Z
CATEGORIES:DepKo
SUMMARY:Depko:An Information Theory Framework for Viewpoint Selection and Mesh Saliency
DESCRIPTION:An Information Theory Framework for Viewpoint Selection and Mesh SaliencyMiquel Feixas, University of Girona\n\nViewpoint selection is an emerging area in computer graphics with applications in fields such as scene exploration, image-based modeling, and volume visualization. Best view selection algorithms are used to obtain the minimum number of views in order to understand or model an object or scene better. In this talk, we present a unified framework for viewpoint selection and mesh saliency computation based on the definition of an information channel between a set of viewpoints (input) and the set of polygons of an object (output). The mutual information of this channel is shown to be a powerful tool to deal with viewpoint selection, viewpoint stability, object exploration, polygonal information and saliency. The information channel is also extended to deal with volume visualization. Finally, we present the application of viewpoint mutual information to mesh simplification.
LOCATION:MRZ Seminar Room
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_301
DTSTAMP:20100614T093403
DTSTART:20100701T080000Z
DTEND:20100701T100000Z
CATEGORIES:DepKo
SUMMARY:Depko: Informational Aesthetic Measures and the Evolution of Van Gogh's Artwork
DESCRIPTION:Informational Aesthetic Measures and the Evolution of Van Gogh's Artwork - Miquel Feixas, University of Girona:\n\nIn 1928, George D. Birkhoff introduced the Aesthetic Measure, defined as the ratio between order and complexity, and, in 1965, Max Bense analyzed Birkhoff’s measure from an information theory point of view. In this talk, the concepts of order and complexity in an image are analyzed in the light of information-theoretic measures and Kolmogorov complexity. The measures proposed are applied to several works. In particular, we analyze the evolution of van Gogh’s style towards the Auvers final period using some informational measures based on the entropy of the palette, the compressibility of the image, an information channel to capture the basic structure of the painting, and other entropy-based measures that take into account spatial information. The values of these measures match fairly well against the different styles in van Gogh’s work and allow a quantitative description of these periods. In addition, we present two measures which quantify the information associated with both color and canvas regions and enable us to visualize the most salient colors and elements of a painting.\n\nHost: Christian Wallraven\n\n
LOCATION:MRZ Seminar Room
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_308
DTSTAMP:20100707T115509
DTSTART:20100712T091500Z
DTEND:20100712T100000Z
CATEGORIES:DepKo
SUMMARY:Depko: Emerging Developments in Cellular MRI
DESCRIPTION:Eric T. Ahrens, Carnegie Mellon University\n\nImaging specific cell populations in vivo can play an important role as a diagnostic, elucidate disease pathogenesis, and accelerate therapeutic development. Recent advances in MRI reagent technologies enable cell tracking and quantification, i.e., cytometry, in vivo. These new methods rely on unique perfluorocarbon (PFC) nanoemulsion formulations. Cells are labeled with PFC either ex vivo following selection, or in vivo, and cell migration patterns are visualized non-invasively using 19F MRI. The key advantage of this method is that the images are extremely selective for the labeled cells, with no background signal from the host's tissues. Importantly, we have shown that the absolute number of labeled cells in regions of interest can be estimated directly from the in vivo images. Data show that PFC nanoemulsions are highly biocompatible and nontoxic. Recent rodent applications using the PFC imaging platform to elucidate inflammatory processes and anti-cancer therapeutic action will be described, as well as the prospects for clinical translation. Additionally, we will describe recent developments in nucleic acid-based MRI reporters of transgene activity in vivo.\n\n
LOCATION:MPH Seminar Room
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_309
DTSTAMP:20100707T120755
DTSTART:20100720T091500Z
DTEND:20100720T100000Z
CATEGORIES:DepKo
SUMMARY:Depko: Benchmarking for research on amyotrophic lateral sclerosis models
DESCRIPTION:\n\nProf. Pavle R. Andjus, Institute for Physiology & Biochemistry, University of Belgrade, Serbia\n\nAmyotrophic lateral sclerosis (ALS) is a devastating neurological disorder affecting upper and lower motoneurons. It is characterized by progressive muscle weakness and atrophy. It is almost invariably fatal, usually within 3 to 6 years after the beginning of the symptoms. There are two types, sporadic and familial that differ in the aetiopathogenesis although they have similar clinical symptoms and neuropathology. Moreover, there is substantial evidence that both types of ALS develop an oxidative stress phase (by extensive glutamate stimulated Ca2+ inflow or by impaired SOD activity) that eventually leads to excitotoxicity. The cause of sporadic ALS is not known but several theories on the mechanism of pathogenesis have been proposed, including a) excitotoxic stimulation due to accumulation of glutamate, and b) immune mechanisms. Our studies in the course of the last decade using different biophysical techniques to tackle the pathophysiology in ALS will be reviewed.\n\nNeuroimmunological study of ALS was performed on hippocampal neurons in culture. Sera from sporadic ALS patients were used to isolate IgGs. Direct ALS IgG effect on Cai2+ homeostasis was measured with confocal microscopy with a fluorescent dye selective for Ca2+ (fluo-3). ALS IgGs specifically blocked, in an irreversible manner, the P/Q type channel activity. An excitotoxic hypothesis was still applicable assuming a suppression of the Cai2+ -dependent desensitization of NMDA receptors. ALS IgGs also induced a delayed Cai2+ transient response to KCl or glutamate in glial-like cells. ALS IgG fraction can indeed affect the presynaptic glutamate release mechanisms as shown in a whole-cell patch clamp study of spontaneous non-NMDA excitatory postsynaptic currents. No changes were observed with IgGs from healthy donors or from patients with Alzheimer’s disease.\n\nRecent results on the model of familial ALS, the transgenic hSOD1 G93A mutant rat demonstrated its feasibity for an in vivo follow-up study. By means of a wide bore 1.5 T MRI instrument and a surface head coil we managed to image degenerative foci (hyperintensity changes) in the brain and brainstem of the model rat. The hyperintensity changes could be observed already in presymptomatic animals but became more apparent in later stages of the disease. Immunocytochemistry of these regions revealed reactive astroglia and activated microglia/macrophages revealing a neuron-glia interaction. In addition, MRI scanning with a 3 T wide bore magnet revealed infiltrations of magnetically labeled T cells in the brainstem. Gd-DTPA contrast revealed a compromised BBB that may serve as the passage for inflammatory immune cells in the vicinity of dilated lateral ventricles.\n\nThe obtained results are in congruence and complementarity to other studies on neuroimmunology and inflammatory processes in ALS. There is still considerable work to be done on the relationship of two forms of ALS, however our studies point that the common denominator may be the function of micro- and macro- glial cells.\n\n\n\nProf. Pavle R. Andjus\n\nInstitute for Physiology & Biochemistry,\n\nSchool of Biology, University of Belgrade\n\nStudentski trg 3\n\n11000 Belgrade, Serbia\n\npandjus@bio.bg.ac.rs
LOCATION:Max Planck Institute for Biological Cybernetics
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_337
DTSTAMP:20100802T091934
DTSTART:20100803T091500Z
DTEND:20100803T100000Z
CATEGORIES:DepKo
SUMMARY:Depko: Heading Discrimination in Head and World Coordinates and Detection of Linear and Angular Components of Curved-path Motion
DESCRIPTION:1) Heading Discrimination in Head and World Coordinates and 2) Detection of Linear and Angular Components of Curved-path Motion \n\n\n\nPaul MacNeilage, Integrated Center for Research and Treatment of Vertigo, Balance, and Ocular Motor Control, University Hospital of Munich\n\n\n\nAbstract: The vestibular cue to heading is the direction of inertial  acceleration that accompanies transient linear movements. This cue is  transduced by the otolith organs. The otoliths also respond to  gravitational acceleration, so vestibular heading discrimination could  depend on: 1) the direction of movement in head coordinates (i.e.,  relative to the otoliths), 2) the direction of movement in world  coordinates (i.e., relative to gravity), or 3) body orientation (i.e.,  the direction of gravity relative to the otoliths). To quantify these  effects, we measured vestibular and visual discrimination of heading  along azimuth and elevation dimensions with observers oriented both  upright and side-down relative to gravity. We compared vestibular  heading thresholds with corresponding measurements of sensitivity to  linear motion along lateral and vertical axes of the head (coarse  direction discrimination and amplitude discrimination). Neither heading  nor coarse direction thresholds depended on movement direction in world  coordinates, demonstrating that the nervous system compensates for  gravity. Instead, they depended similarly on movement direction in head  coordinates (better performance in the horizontal plane), and on body  orientation (better performance in the upright orientation). Heading  thresholds were correlated with, but significantly larger than,  predictions based on sensitivity in the coarse discrimination task. Perceptual interactions between otolith and horizontal semicircular  canal signals were examined using a classical signal detection paradigm.  In a rotation detection experiment, blindfolded subjects were asked to  detect the presence of angular motion in blocks where half of the trials  were pure naso-occipital translation and half were simultaneous  translation and yaw rotation (curved-path motion). In separate,  translation detection experiments, subjects were also asked to detect  the presence or absence of naso-occipital linear motion in blocks where  half of the trials were pure yaw rotation and half were curved path.  Rotation thresholds increased slightly but not significantly with  concurrent linear velocity magnitude. Translation thresholds, on the  other hand, increased significantly with increasing magnitude of  concurrent angular velocity. These findings suggest that conscious  perception might not have independent access to separate estimates of  linear and angular movement parameters during curved-path motion.  Estimates of linear (and perhaps angular) components might instead rely  on integrated information from canals and otoliths. Such interaction may  underlie previously reported perceptual errors during curved-path  motion, and may originate from mechanisms that are specialized for  tilt-translation processing during vertical plane rotation.
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Michael Barnett-Cowan":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_346
DTSTAMP:20100930T085119
DTSTART:20101014T120000Z
DTEND:20101014T130000Z
CATEGORIES:DepKo
SUMMARY:[Friko]: When We Get "When?" Wrong
DESCRIPTION:When We Get "When?" Wrong - Chris Cantor, University of Berkeley, USA\n\n\n\nPsychophysicists are often faced with the problem of measuring\n\nobservers’ experiences of time-varying stimuli.  A common strategy is\n\nto use an ultra-brief visual event as a temporal marker and have\n\nobservers render a decision based on what they saw at "that moment in\n\ntime".  Using this paradigm may not always be a good idea.  Under\n\ncertain conditions, human observers can judge the temporal order of\n\nbrief visual events with high accuracy, correctly distinguishing\n\nstimuli separated by several milliseconds  (e.g. Westheimer & McKee,\n\n1977).  On the other hand, human observers also demonstrate biases in\n\nwhich they misjudge the order of brief visual events separated by as\n\nmuch as 100 ms (e.g. Cantor & Schor, 2009).    We will discuss\n\npsychophysical data and modeling suggesting that early visual\n\nprocessing profoundly alters the temporal structure of the visual\n\ninput and that this distortion is accessible to conscious perception.\n\n\n\n\n\nHost: Jess Hartcher O'Brien\n\n\n\n
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Jess Hartcher-O'brien":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_353
DTSTAMP:20101103T134219
DTSTART:20101126T101500Z
DTEND:20101126T111500Z
CATEGORIES:DepKo
SUMMARY:Depko: Information-theoretic computer vision for autonomous robots
DESCRIPTION:Boyan Bonev (University of Alicante, Spain)\n\n\n\nComputer  vision is one of the richest sources of information for an autonomous  robot. Mobile robots are getting out of the laboratories as they get  closer to humans. They have to deal with unconstrained environments and  perform a wide range of tasks. Localization, navigation, recognition and  co-operation are some of the basic skills which form autonomy. In this  challenging scenario, computer vision has became a major target of  research. It sets out the need for paradigms which optimize the  extraction of useful information from images.  Information theory sets the basics of many pattern recognition problems  related to computer vision. It provides tools, not only for recognizing  patterns in single images and performing classifications tasks, but also  for dealing with sequences of images, like image alignment,  segmentation of a sequence of frames and selection of features for  coding a set of images. All these are necessary for building a map, for  localization, as well as for providing scalability to the approach. In this talk we will discuss these topics and we will present examples  of application to different robots: quadruped, wheeled, as well as  quadcopters. We will talk about the computational complexity problems  which arise with information-theoretic approaches and how can we deal with them.
LOCATION:Max Planck House Lecture Hall
ORGANIZER;CN="Paolo Robuffo-Giordano":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_363
DTSTAMP:20101203T135506
DTSTART:20110104T101500Z
DTEND:20110104T111500Z
CATEGORIES:DepKo
SUMMARY:[Depko] Micro-Valences: Affective valence in "neutral" object perception
LOCATION:Max Planck House Lecture Hall
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_405
DTSTAMP:20110207T095541
DTSTART:20110210T140000Z
DTEND:20110210T150000Z
CATEGORIES:DepKo
SUMMARY:Depko: Building Autonomous Humanoids
DESCRIPTION:Building humanoid robots able to learn to operate in the real world and to interact and communicate with humans, must model and reflectively reason about their perceptions and actions in order to learn, act, predict and react appropriately. Such capabilities can only be attained through physical interaction with and exploration of the real world and requires the simultaneous consideration of perception and action. Representations built from such interactions are much better adapted to guiding behaviour than human crafted rules and allow situated and embodied systems, such as humanoid robots in human-centered environments, to gradually extend their cognitive horizon. To achieve this goal I am building humanoid robots with complex and rich sensorimotor capabilities as the most suitable experimental platform for studying cognitive information processing.\n\n\n\nIn this talk, I will present recent progress towards building autonomous humanoid robots able to act, interact in and autonomously acquire knowledge in the real world. Apart from their design and mechatronics, current results will be presented towards the implementation of integrated 24/7 humanoid robots able to 1) perform complex grasping and manipulation tasks in a kitchen environment 2) autonomously acquire object knowledge through active visual and haptic exploration and 3) learn actions from human observation and imitate them in goal-directed manner. The developed capabilities will be demonstrated on the humanoid robots ARMAR-IIIa and ARMAR-IIIb.\n\n\n\n\n\nBIO:\n\nTamim Asfour is senior research scientist and group leader of the Humanoid Research Group at Humanoids and Intelligence Systems Lab, Institute for Anthropomatics, Karlsruhe Institute of Technology (KIT). He received his diploma degree in Electrical Engineering (Dipl.-Ing.) in 1994 and his PhD in Computer Science (Dr.-Ing.) in 2003 from the University of Karlsruhe. In 2003 he was awarded with the Research Center for Information Technology (FZI) price for his Ph.D. thesis on sensorimotor control in humanoid robotics and the development of the humanoid robot ARMAR. Since September 2010 he holds an Adjunct Professor position at the Georgia Institute of Technology (Georgia Tech), College of Computing, School of Interactive Computing.\n\n\n\nTamim Asfour is member of the Editorial Board of IEEE Transactions on Robotics and European Chair of the IEEE-RAS Technical Committee on Humanoid Robots. He is member the Executive Board of the German Association of Robotics (DGR: Deutsche Gesellschaft für Robotik). He serves as member on several program committees and review panels.\n\n\n\nHis major research interest is humanoid robotics. In particular, his research topics include action learning from human observation, goal-directed imitation learning, dexterous grasping and manipulation, active vision and active touch, whole-body motion planning, cognitive control architectures, system integration, robot software and hardware control architecture, motor control and mechatronics.
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Jan Peters":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_406
DTSTAMP:20110207T100135
DTSTART:20110223T110000Z
DTEND:20110223T120000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGLO]: Properties of the canonical inter-areal circuit
DESCRIPTION:We analyze the patterns of inter-areal connectivity as obtained from injections of retrograde tracers into macaque cortex, based on indices related to the strength of connection and its laminar distribution. The network is sufficiently dense that the distinction between a small world and a random network disappears.  Nevertheless, we have found certain canonical properties in the patterns of inter-areal connectivity.  First, the distribution of the weights of the afferent projections follows a log normal distribution  that spans at least 5 orders of magnitude.  In addition, the strength of the weights depends exponentially on the distance of the projection.  We have shown that these two properties are linked by the distribution of inter-areal distances between areas, which follows a truncated normal distribution. The exponential distance rule (EDR) of weights confers high local and global efficiency on the transfer of information in a way that a random distance rule does not.  The global efficiency depends on the strong, short-distance projections because it is relatively unaffected by elimination of the weak, long-distance pathways while the local efficiency is initially increased by the elimination of the weak, long-distance pathways.  Thus, the small world-like properties that differentiate our dense graph from a random one emerge from the distribution of the weights of the projections and not from the binary topology.
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Matthias Munk":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_414
DTSTAMP:20110310T150245
DTSTART:20110314T140000Z
DTEND:20110314T150000Z
CATEGORIES:DepKo
SUMMARY:[Depko]: A Cybernetic Approach to Assess Flight Simulator Motion Fidelity
DESCRIPTION:Compared to aircraft, flight simulators are severely limited in their motion envelope. This causes the generation of motion cues in flight simulation to be an inevitable\n\ncompromise between the desired level of motion cue fidelity and the available motion\n\nspace. Due to a limited understanding of human multimodal motion perception during aircraft control, the definition of requirements for flight simulator motion fidelity\n\nis a problem the flight simulation community has struggled with for years. The development of adequate tuning procedures for motion washout algorithms has remained\n\nproblematic for the same reason.\n\n\n\nThis presentation will give an overview of a research project that was initiated at\n\nDelft University of Technology in 2006, which aims to develop an objective method for\n\nthe assessment of simulator motion fidelity. This proposed method makes use of a cybernetic approach, based on the capturing of pilot’s responses to visual and (simulator)\n\nmotion cues using multimodal pilot models, to obtain quantitative measurements and\n\nincreased understanding of how these different modalities interact during manual aircraft control. Furthermore, by fitting such multimodal pilot models to measurements\n\nof pilot control behavior taken both in the full motion flight simulator and the laboratory aircraft of Delft University of Technology, an attempt is made to objectively\n\nmeasure the behavioral discrepancies that may occur in flight simulators compared to\n\nreal flight, and to relate these observed discrepancies to the choices in the presentation\n\nof simulator motion cues.\n\n\n\nA recently collected set of such in-flight and simulator measurements for aircraft\n\npitch and roll attitude control tasks shows that this approach indeed allows for objective\n\ndetection of changes in pilot control that result from degraded simulator motion fidelity.\n\nTrue evaluation of the effects of degraded motion fidelity through comparison of in-\n\nflight measurements of pilot behavior and those taken in a flight simulator is, however,\n\nfound to remain difficult, as other factors than the differences in the supplied motion\n\ncues are also found to have a pronounced influence on the control strategy adopted by\n\npilots.
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Frank Nieuwenhuizen":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_410
DTSTAMP:20110307T160023
DTSTART:20110315T101500Z
DTEND:20110315T110000Z
CATEGORIES:DepKo
SUMMARY:[Depko]: The perception of movies
DESCRIPTION:We all watch movies, and are absorbed into them to the extent that we forget that we are mere observers of flickering images on a screen. What is it about watching movies that gives us the illusion of presence, and what are the implications for theories of visual perception? A naive approach might predict that a visual environment containing frequent temporal discontinuities would be aversive, but this is clearly not the case in movies. Why not? I will consider the basis of the feeling of presence, and describe experiments aimed at establishing the effect of various optical and temporal factors which affect presence during the viewing of movies.
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Heinrich H. Bülthoff":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_417
DTSTAMP:20110318T155516
DTSTART:20110322T090000Z
DTEND:20110322T100000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGMB]: Toward Autonomous Visual Learning
DESCRIPTION:Vision is a crucial sense for computational systems to interact with their environments as biological systems do. A major task is interpreting images of complex scenes, by recognizing and localizing objects, persons and actions. This involves learning a large number of visual models, ideally autonomously. In this talk I will present two ways of reducing the amount of human supervision required by this learning process.\n\n\n\nThe first way is labeling images only by the object class they contain. Learning from cluttered images is very challenging in this weakly supervised setting. In the traditional paradigm, each class is learned starting from scratch. In our work instead, knowledge generic over classes is first learned during a meta-training stage from images of diverse classes with given object locations, and is then used to support learning any new class without location annotation. Generic knowledge helps because during meta-training the system can learn about localizing objects in general. As demonstrated experimentally, this approach enables learning from more challenging images than possible before, such as the PASCAL VOC 2007, containing extensive clutter and large scale and appearance variations between object instances.\n\n\n\nThe second way is the analysis of news items consisting of images and text captions. We associate names and action verbs in the captions to the face and body pose of the persons in the images. We introduce a joint probabilistic model for simultaneously recovering image-caption correspondences and learning appearance models for the face and pose classes occurring in the corpus. As demonstrated experimentally, this joint `face and pose' model solves the correspondence problem better than earlier models covering only the face.\n\n\n\nI will conclude with an outlook on the idea of visual culture, where new visual concepts are learned incrementally on top of all visual knowledge acquired so far. Beside generic knowledge, visual culture includes also knowledge specific to a class, knowledge of scene structures and other forms of visual knowledge. Potentially, this approach could considerably extend current visual recognition capabilities and produce an integrated body of visual knowledge.\n\n\n\nIf you would like to meet with Dr. Ferrari, please \n\ncontact Melanie Feldhofer  .\n\n\n\nVittorio Ferrari\n\nETH Zurich\n\nhttp://www.vision.ee.ethz.ch/~calvin/
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Michael Black":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_420
DTSTAMP:20110401T090057
DTSTART:20110405T091500Z
DTEND:20110405T100000Z
CATEGORIES:DepKo
SUMMARY:[Depko]: Integrating multisensory information for motor learning
DESCRIPTION:\n\nDenise Henriques - York University, Canada\n\n\n\nIn this talk, I will discuss recent studies from my lab investigating how the brain\n\nuses information from one sense to recalibrate other senses in order to produce\n\nconsistent representations for motor control. More specifically, this research\n\nasks whether recalibrating the arm motor system based on information from\n\none sense, like vision, also recalibrates other senses such as proprioception.\n\nFrom previous research, we know that when subjects reach to visual targets\n\nwith altered visual feedback of the hand, subjects learn to adjust their hand\n\nmovements in order to get the hand (or cursor) onto the target. We have\n\nrecently shown that this visuomotor adaptation also affects subjects’ perceptual\n\njudgments regarding the location of their unseen hand position, even when\n\nestimating hand position after its actively versus passively moved. That is,\n\nlearning to reach with altered visual feedback of the hand not only leads to\n\nmotor recalibration, but also partial proprioceptive recalibration of the hand in\n\nthe direction of the visuomotor distortion. We have found that proprioception\n\nis recalibrated to a similar extent across the life-span and for patients with\n\nParkinson diseases. The extent of this recalibration is independent of the degree\n\nof proprioceptive sensitivity. Moreover, we have found that mere exposure to a\n\nsensory discrepancy between the seen and felt hand position when the hand is\n\npassively moved also leads to motor and proprioceptive recalibration of the hand.\n\nTaken together, our results suggest that sensorimotor learning affects not only\n\nmotor output and the predicted sensory consequences but also affects non-\n\nvisual estimates of static hand position. In this talk, I will show evidence that\n\nsuggests that these motor and sensory recalibrations (or plasticity) are separate\n\nprocesses, which rely on different error signals, including sensory discrepancies\n\nbetween visual and proprioceptive estimates of hand position. Specifically, our\n\nresults indicate that the patterns of motor and sensory recalibration tend to differ\n\nfor bimanual transfer and generalization across the workspace. These studies\n\nexpand our understanding of the interaction of vision and body-position sense in\n\nshaping the brain’s motor-control systems, as well as provide insight into the type\n\nof error signals and mechanisms that may underlie learning.
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Michael Barnett-Cowan":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_421
DTSTAMP:20110401T090249
DTSTART:20110407T091500Z
DTEND:20110407T100000Z
CATEGORIES:DepKo
SUMMARY:[Depko]: iMOT: interactive Multiple Object Tracking
DESCRIPTION:\n\nSpeaker: Ian Thornton - University of Swansea, UK\n\n\n\nIn a standard Multiple Object Tracking (MOT) task, participants passively track a set of independently moving objects. Performance on this task has been used to shed light on the nature, capacity limits and control of attention. In this talk I will describe a new iPad task in which participants interact with objects, rather than simply track them. To do this we combined a typical MOT display with the path-guidance, touch-screen interface popularised in smart phone games such as Flight Controller and Harbor Master. I will report basic behavioural data (number of collisions, number of paths, etc.,) from a series of experiments in which the set-size and/or speed of items was varied. I will also discuss how accelerometer data from the iPad can be used as an additional measure of performance.
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Prof. Dr. Heinrich H. Bülthoff":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_411
DTSTAMP:20110309T190008
DTSTART:20110418T120000Z
DTEND:20110418T130000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGMB]: Learning Representations for Real-world Recognition
DESCRIPTION:Speaker: Trevor Darrell (UC Berkeley, USA)\n\nMethods for visual recognition have made dramatic strides in recentyears on various online benchmarks, but performance in the real worldstill often falters.  Classic gradient-histogram models make overlysimplistic assumptions regarding image appearance statistics, bothlocally and globally.  Recent progress suggests that newlearning-based representations can improve recognition by devices thatare embedded in a physical world.  I'll review new methods for domainadaptation which capture the visual domain shift between environments,and improve recognition of objects in specific places when trainedfrom generic online sources.  I'll discuss methods for cross-modalsemi-supervised learning, which can leverage additional unlabeledmodalities in a test environment. Finally as time permits I'll presentrecent results learning hierarchical local image representations basedon recursive probabilistic topic models, on learning strong objectcolor models from sets of uncalibrated views using a new multi-viewcolor constancy paradigm, and/or on recent results on monocularestimation of grasp affordances.Prof. Trevor Darrell is on the faculty of the UC Berkeley EECSdepartment, and leads the computer vision group at the InternationalComputer Science Institute in Berkeley, CA.  He obtained the PhD fromMIT in 1996, and worked at Interval Research in Palo Alto from 1996 to1999. From 1999 to 2008 he served on the faculty of the MIT EECSdepartment, where he led the CSAIL Vision Interfaces group. In 2008 hereturned to the west coast, and since then has been pursuing hisinterests in visual perception, machine learning, multimodalinterfaces, and entrepreneurial activity in Berkeley.  Prof. Darrellhas authored over 100 journal and conference papers, has served on theeditorial boards of PAMI and the Artificial Intelligence Journal, andserved as Program Chair for CVPR 2010.  He also serves on the advisoryboards of local start-ups IQ Engines and BotSquare.
LOCATION:MPH Lecture Hall
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_426
DTSTAMP:20110418T152921
DTSTART:20110419T120000Z
DTEND:20110419T130000Z
CATEGORIES:DepKo
SUMMARY:[Depko]: Tracking and measurement in sports video
DESCRIPTION:Speaker: Jim Little, University of British Columbia\n\nAbstract:I  will survey our work on tracking and measurement, waypoints on the path to activity recognition and understanding, in sports video, highlighting some of our recent work on rectification and player tracking, not just in hockey but more recently in basketball, where we have addressed player identification both in a fully supervised and semi-supervised manner.\n\nLink: http://www.cs.ubc.co/~little/\n\n\n\n
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Michael Black":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_479
DTSTAMP:20110506T115939
DTSTART:20110513T121500Z
DTEND:20110513T130000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGMB]: Capturing and Synthesizing Hand-Object Interaction 
DESCRIPTION:Henning Hamer\n\nETH Zurich\n\nhttp://www.vision.ee.ethz.ch/members/get_member.cgi?lang=de&id=149\n\n\n\nThe scope of this work is hand-object interaction. As a starting point, we observe hands manipulating objects and derive information based on computer vision methods. After considering hands and objects in isolation, we focus on the inherent interdependencies. One application of the gained knowledge is the synthesis of interactive hand motion for animated sequences.\n\n\n\nIf you would like to meet with Henning Hamer, please contact Melanie Feldhofer
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Michael Black":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_540
DTSTAMP:20110523T080128
DTSTART:20110523T083000Z
DTEND:20110523T093000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGMB]: Emerging Neurotechnologies for Interfacing the Brain
DESCRIPTION:Speaker: Arto V. Nurmikko\n\nSchool of Engineering and Brain Sciences Institute, Brown University\n\nhttp://www.engin.brown.edu/people/Faculty/facultypage.php?id=1106970197\n\n\n\nInvasive access by microprobe arrays inserted safely into the brain is now enabling us to “listen” to local neural circuits at levels of spatial and temporal detail which, in addition to enriching fundamental brain science, has led to the possibility of a new generation of neurotechnologies to overcome disabilities due to a range of neurological injuries where pathways from the brain to the rest of the central and peripheral nervous systems have been injured or severed. In this presentation we discuss the biomedical engineering challenges and opportunities with these incipient technologies, with emphasis on implantable wireless neural interfaces for communicating with the brain. A second topic, related to the possibility of sending direct inputs of information back to the brain by implanted devices is also explored, focusing on recently discovered means to render selected neural cell types and microcircuits to be light sensitized, following local microbiologically induced conditioning.\n\nIf you would like to meet with Arto Nurmikko, please contact Melanie Feldhofer: melanie.feldhofer@tuebingen.mpg.de
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_541
DTSTAMP:20110523T080139
DTSTART:20110523T083000Z
DTEND:20110523T093000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGMB]: Emerging Neurotechnologies for Interfacing the Brain
DESCRIPTION:Speaker: Arto V. Nurmikko\n\nSchool of Engineering and Brain Sciences Institute, Brown University\n\nhttp://www.engin.brown.edu/people/Faculty/facultypage.php?id=1106970197\n\n\n\nInvasive access by microprobe arrays inserted safely into the brain is now enabling us to “listen” to local neural circuits at levels of spatial and temporal detail which, in addition to enriching fundamental brain science, has led to the possibility of a new generation of neurotechnologies to overcome disabilities due to a range of neurological injuries where pathways from the brain to the rest of the central and peripheral nervous systems have been injured or severed. In this presentation we discuss the biomedical engineering challenges and opportunities with these incipient technologies, with emphasis on implantable wireless neural interfaces for communicating with the brain. A second topic, related to the possibility of sending direct inputs of information back to the brain by implanted devices is also explored, focusing on recently discovered means to render selected neural cell types and microcircuits to be light sensitized, following local microbiologically induced conditioning.\n\nIf you would like to meet with Arto Nurmikko, please contact Melanie Feldhofer: melanie.feldhofer@tuebingen.mpg.de
LOCATION:MPH Lecture Hall
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_542
DTSTAMP:20110523T080148
DTSTART:20110523T083000Z
DTEND:20110523T093000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGMB]: Emerging Neurotechnologies for Interfacing the Brain
DESCRIPTION:Speaker: Arto V. Nurmikko\n\nSchool of Engineering and Brain Sciences Institute, Brown University\n\nhttp://www.engin.brown.edu/people/Faculty/facultypage.php?id=1106970197\n\n\n\nInvasive access by microprobe arrays inserted safely into the brain is now enabling us to “listen” to local neural circuits at levels of spatial and temporal detail which, in addition to enriching fundamental brain science, has led to the possibility of a new generation of neurotechnologies to overcome disabilities due to a range of neurological injuries where pathways from the brain to the rest of the central and peripheral nervous systems have been injured or severed. In this presentation we discuss the biomedical engineering challenges and opportunities with these incipient technologies, with emphasis on implantable wireless neural interfaces for communicating with the brain. A second topic, related to the possibility of sending direct inputs of information back to the brain by implanted devices is also explored, focusing on recently discovered means to render selected neural cell types and microcircuits to be light sensitized, following local microbiologically induced conditioning.\n\nIf you would like to meet with Arto Nurmikko, please contact Melanie Feldhofer: melanie.feldhofer@tuebingen.mpg.de
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Prof. Michael Black":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_543
DTSTAMP:20110523T080149
DTSTART:20110523T083000Z
DTEND:20110523T093000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGMB]: Emerging Neurotechnologies for Interfacing the Brain
DESCRIPTION:Speaker: Arto V. Nurmikko\n\nSchool of Engineering and Brain Sciences Institute, Brown University\n\nhttp://www.engin.brown.edu/people/Faculty/facultypage.php?id=1106970197\n\n\n\nInvasive access by microprobe arrays inserted safely into the brain is now enabling us to “listen” to local neural circuits at levels of spatial and temporal detail which, in addition to enriching fundamental brain science, has led to the possibility of a new generation of neurotechnologies to overcome disabilities due to a range of neurological injuries where pathways from the brain to the rest of the central and peripheral nervous systems have been injured or severed. In this presentation we discuss the biomedical engineering challenges and opportunities with these incipient technologies, with emphasis on implantable wireless neural interfaces for communicating with the brain. A second topic, related to the possibility of sending direct inputs of information back to the brain by implanted devices is also explored, focusing on recently discovered means to render selected neural cell types and microcircuits to be light sensitized, following local microbiologically induced conditioning.\n\nIf you would like to meet with Arto Nurmikko, please contact Melanie Feldhofer: melanie.feldhofer@tuebingen.mpg.de
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Prof. Michael Black":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_545
DTSTAMP:20110523T093632
DTSTART:20110526T120000Z
DTEND:20110526T130000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGBU]: Haptic Shared Control: designing intuitive Human-Machine interaction based on visual and neuromuscular modeling
DESCRIPTION:Dr. ir. David Abbink, Delft University of Technology, The Netherlands\n\n\n\nHaptic shared control is a promising way to support humans through forces on the control interface when controlling vehicles or during remote manipulation. Although many papers have shown beneficial results compared to manual control or full automation, the design of the feedback forces remains a trial-and-error process. It is recognized in literature that the human response to forces can change substantially, and although much relevant knowledge about measuring and modeling such adaptations exists in the field of neuroscience, the haptic community has not employed it for shared control design. \n\nIn this presentation I will give a brief overview of the current state of the art in haptic shared control for several applications, and argue the importance of measuring neuromuscular properties as a basis for the design of shared control forces. If designed correctly, haptic shared control can serve as a smooth grading between manual control and full automation. To illustrate all of this, several case studies will be discussed, ranging from my work for Nissan and Boeing to ongoing work in remote manipulation. \n\nBIO:\n\nDavid Abbink (1977) received his MSc degree in Mechanical Engineering at Delft University of Technology in 2002. Subsequently he worked for three years on a research project funded by Nissan, during which he helped develop and evaluate a force feedback gas pedal to support driver with car following, which was released as "Distance Control Assist" on American and Japanese markets in 2008. \n\nIn 2006 he received his PhD degree as a result of his thesis "Neuromuscular Analysis of Haptic Feedback during Car Following", which was awarded that year's best PhD thesis in the area of movement sciences in the Netherlands. He continued his research on haptics and neuromuscular analysis during several projects for Nissan and Boeing, and received a position as Assistant Professor at the Delft Haptics Lab in 2009. Shortly after he was awarded the Dutch "VENI grant" to further stimulate his work on the design of human-centered haptic guidance. \n\nHe co-wrote the "H-Haptics" project which was granted in May 2011, yielding funding for 16 PhDs to extend the concept of haptic shared control from vehicular control to tele-operation in domains such as nuclear remote handling, and manipulation in deep sea or space.\n\n\n\n
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Joost Venrooij":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_647
DTSTAMP:20110530T081913
DTSTART:20110601T121500Z
DTEND:20110601T130000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGBU]: High-throughput electron microscopy of mouse cortex for 3D reconstruction
DESCRIPTION:\n\nSpeaker: Daniel Berger – MIT, USA\n\nHigh-throughput electron microscopy of mouse cortex for 3D reconstruction\n\nThe high speed and storage capacity of modern computers make it possible to store and process data sets in the Terabyte range. Neuroscientists can now work with stacks of electron microscopic images of nervous tissue which are large and detailed enough to densely reconstruct axons and dendrites, and synapses between them. This opens a fascinating new field of neuroscience research, called 'Connectomics'.\n\nThis work poses a number of challenges, and many are of a computational nature. On one side, the data acquisition should be automated and made more robust to allow for the lossless collection of larger image stacks. On the other, the size of these data sets makes image processing difficult and complete manual segmentation unfeasible.\n\nSeveral methods have been developed to automate cutting and imaging, most prominently serial block-face scanning electron microscopy (SBF-SEM), focused ion beam milling (FIB-SEM), and automatically collecting slices from a standard ultramicrotome on tape (ATUM-SEM). In collaboration with the lab of Jeff Lichtman at Harvard, we work with stacks of ATUM-SEM images of mouse cortex. First the individual SEM images have to be aligned and stitched to yield a consistent three-dimensional image block. We then segment smaller sub-volumes manually (on the order of 10 micrometers cubed). This already enables us to compute statistics on synapse size, local connectivity and the relative volume of different neurite types. We then use these manual segmentations to train machine-learning algorithms with the goal to automate image segmentation. The algorithm generates a candidate segmentation which is then viewed and corrected by human experts.
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Heinrich H. Bülthoff":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_648
DTSTAMP:20110530T082154
DTSTART:20110606T091500Z
DTEND:20110606T100000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGBU]: Computational and Neuroimaging Approaches to Perceiving Object Properties
DESCRIPTION:\n\n\n\nSpeaker: Dan Kersten – University of Minnesota, USA\n\n\n\nComputational and Neuroimaging Approaches to Perceiving Object Properties\n\nThe visual system of the brain is believed to be composed of a hierarchy of cortical areas in which simple local features of an object, such as its edges, are first extracted in primary visual cortex (V1) and then passed on to higher areas to extract more complex properties, such as size and shape. This model has been extraordinarily fruitful, providing explanations of a considerable body of psychophysical and neurophysiological results. However, there is growing evidence that primary visual cortex is doing a whole lot more than extracting local features. I will describe brain imaging studies in humans showing that: 1) The spatial extent of activity in primary visual cortex is affected by the perceived size of an object, suggesting a top-down flow of depth information to V1; and 2) Localized BOLD activity in V1 is responsive to the apparent lightness of an object's surface, even in the absence of a local change in light level.  I will discuss these results in terms of Bayesian theories of vision that involve the tight interplay of feedforward, lateral, and feedback processing.\n\nThe visual system of the brain is believed to be composed of a hierarchy of cortical areas in which simple local features of an object, such as its edges, are first extracted in primary visual cortex (V1) and then passed on to higher areas to extract more complex properties, such as size and shape. This model has been extraordinarily fruitful, providing explanations of a considerable body of psychophysical and neurophysiological results. However, there is growing evidence that primary visual cortex is doing a whole lot more than extracting local features. I will describe brain imaging studies in humans showing that: 1) The spatial extent of activity in primary visual cortex is affected by the perceived size of an object, suggesting a top-down flow of depth information to V1; and 2) Localized BOLD activity in V1 is responsive to the apparent lightness of an object's surface, even in the absence of a local change in light level.  I will discuss these results in terms of Bayesian theories of vision that involve the tight interplay of feedforward, lateral, and feedback processing.
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Heinrich H. Bülthoff":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_656
DTSTAMP:20110614T073401
DTSTART:20110614T080000Z
DTEND:20110614T090000Z
CATEGORIES:DepKo
SUMMARY:[Depko]: Increasing safety of helicopter operations
DESCRIPTION:Speaker:  - Dr. Marilena D. Pavel, Delft University, Netherlands\n\n\n\nHelicopters are generally reliable flying machines capable of fulfilling missions impossible \n\nwith fixed-wing aircraft, most notably rescue operations. These missions, however, often lead \n\nto high and sometimes excessive pilot workload. Although high standards in terms of safety \n\nare imposed in helicopter design, studies show that, per flight hour, “it is ten times more \n\nlikely to be involved in an accident in a helicopter than in a fixed-wing aircraft”. Nearly \n\n45 percent of all accidents of single-piston helicopters is attributed to pilot loss of control, \n\nwhere because of various causes, often involving vibrations, high workload, and bad weather, \n\na pilot loses control of the helicopter and crashes, sometimes with fatal consequences. \n\nQuoting the Royal Netherlands Air Force: “for helicopters there is a considerable number of \n\ninexplicable incidents (…) which involved piloting loss of control”. The main cause for the \n\npilot loss of control accidents points out deficiencies in designing helicopter handling \n\nqualities (HQs). \n\n\n\nThe presentation is intended to highlight some recent research activities at Delft University \n\nfor designing of high-performance safer helicopters, that are easier to fly, and capable of \n\nconducting more challenging operations. Four objectives will be discussed: \n\n(1) Development of advanced helicopter flight mechanics models that allow an \n\ninvestigation of the intricate relationship between helicopter vibratory loads and bio- \n\ndynamical interaction effects, often the root of pilot-induced oscillations; \n\n(2) Development of new complementary helicopter HQs criteria for assessing \n\nsimultaneously helicopter performance and vibratory loads. This novel approach for \n\nassessing helicopter HQs can be regarded as a crucial step in extending the handling \n\nqualities standard ADS-33 towards high-performance helicopter operations; \n\n(3) Development of flight simulator guidelines giving how the simulator visual, \n\nmotion, and force-cueing apparatus need to be tuned in order to properly induce pilot \n\ncontrol behaviour in situations with high-vibratory loads; \n\n(4) Development of protocols for training helicopter pilots, including pilot \n\ninstructions, task definitions and mission performance evaluations that can be used \n\nfor more effective pilot training in potentially disastrous situations.\n\n\n\n
LOCATION:MRZ Seminar Room
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_657
DTSTAMP:20110614T073410
DTSTART:20110614T080000Z
DTEND:20110614T090000Z
CATEGORIES:DepKo
SUMMARY:[Depko]: Increasing safety of helicopter operations
DESCRIPTION:Speaker:  - Dr. Marilena D. Pavel, Delft University, Netherlands\n\n\n\nHelicopters are generally reliable flying machines capable of fulfilling missions impossible \n\nwith fixed-wing aircraft, most notably rescue operations. These missions, however, often lead \n\nto high and sometimes excessive pilot workload. Although high standards in terms of safety \n\nare imposed in helicopter design, studies show that, per flight hour, “it is ten times more \n\nlikely to be involved in an accident in a helicopter than in a fixed-wing aircraft”. Nearly \n\n45 percent of all accidents of single-piston helicopters is attributed to pilot loss of control, \n\nwhere because of various causes, often involving vibrations, high workload, and bad weather, \n\na pilot loses control of the helicopter and crashes, sometimes with fatal consequences. \n\nQuoting the Royal Netherlands Air Force: “for helicopters there is a considerable number of \n\ninexplicable incidents (…) which involved piloting loss of control”. The main cause for the \n\npilot loss of control accidents points out deficiencies in designing helicopter handling \n\nqualities (HQs). \n\n\n\nThe presentation is intended to highlight some recent research activities at Delft University \n\nfor designing of high-performance safer helicopters, that are easier to fly, and capable of \n\nconducting more challenging operations. Four objectives will be discussed: \n\n(1) Development of advanced helicopter flight mechanics models that allow an \n\ninvestigation of the intricate relationship between helicopter vibratory loads and bio- \n\ndynamical interaction effects, often the root of pilot-induced oscillations; \n\n(2) Development of new complementary helicopter HQs criteria for assessing \n\nsimultaneously helicopter performance and vibratory loads. This novel approach for \n\nassessing helicopter HQs can be regarded as a crucial step in extending the handling \n\nqualities standard ADS-33 towards high-performance helicopter operations; \n\n(3) Development of flight simulator guidelines giving how the simulator visual, \n\nmotion, and force-cueing apparatus need to be tuned in order to properly induce pilot \n\ncontrol behaviour in situations with high-vibratory loads; \n\n(4) Development of protocols for training helicopter pilots, including pilot \n\ninstructions, task definitions and mission performance evaluations that can be used \n\nfor more effective pilot training in potentially disastrous situations.\n\n\n\n
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Joost Venrooij":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_658
DTSTAMP:20110614T073526
DTSTART:20110614T121500Z
DTEND:20110614T130000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGMB]: Learning Components for Human Sensing
DESCRIPTION:Speaker: Fernando De La Torre (CMU, USA)\n\n\n\nEnabling computers to understand human behavior has the potential to revolutionize many areas that benefit society such as clinical diagnosis, human computer interaction, and social robotics. A critical element in the design of any behavioral sensing system is to find a good representation of the data for encoding, segmenting, classifying and predicting subtle human behavior. In this talk I will propose several extensions of Component Analysis (CA) techniques (e.g., kernel principal component analysis, support vector machines, spectral clustering) that are able to learn spatio-temporal representations or components useful in many human sensing tasks.\n\nIn the first part of the talk I will give an overview of several ongoing projects in the CMU Human Sensing Laboratory, including our current work on depression assessment from video, as well as hot-flash detection from wearable sensors. In the second part of the talk I will show how several extensions of the CA methods outperform state-of-the-art algorithms in problems such as temporal alignment of human motion, temporal segmentation/clustering of human activities, joint segmentation and classification of human behavior, facial expression analysis, and facial feature detection in images. The talk will be adaptive, and I will discuss the topics of major interest to the audience.\n\nhttp://www.cs.cmu.edu/~ftorre/\n\n\n\n\n\n\n\n
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_659
DTSTAMP:20110614T073541
DTSTART:20110614T121500Z
DTEND:20110614T130000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGMB]: Learning Components for Human Sensing
DESCRIPTION:Speaker: Fernando De La Torre (CMU, USA)\n\n\n\nEnabling computers to understand human behavior has the potential to revolutionize many areas that benefit society such as clinical diagnosis, human computer interaction, and social robotics. A critical element in the design of any behavioral sensing system is to find a good representation of the data for encoding, segmenting, classifying and predicting subtle human behavior. In this talk I will propose several extensions of Component Analysis (CA) techniques (e.g., kernel principal component analysis, support vector machines, spectral clustering) that are able to learn spatio-temporal representations or components useful in many human sensing tasks.\n\nIn the first part of the talk I will give an overview of several ongoing projects in the CMU Human Sensing Laboratory, including our current work on depression assessment from video, as well as hot-flash detection from wearable sensors. In the second part of the talk I will show how several extensions of the CA methods outperform state-of-the-art algorithms in problems such as temporal alignment of human motion, temporal segmentation/clustering of human activities, joint segmentation and classification of human behavior, facial expression analysis, and facial feature detection in images. The talk will be adaptive, and I will discuss the topics of major interest to the audience.\n\nhttp://www.cs.cmu.edu/~ftorre/\n\n\n\n\n\n\n\n
LOCATION:MRZ Seminar Room
ORGANIZER;CN="MIchael Black":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_671
DTSTAMP:20110718T071955
DTSTART:20110718T091500Z
DTEND:20110718T100000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGMB]: 3D Isometric Shape Correspondence
DESCRIPTION:\n\nSpeaker: Yusuf Sahillioglu - Koç University, İstanbul/Turkey\n\n3D Isometric Shape Correspondence\n\nAbstract:\n\n3D shape correspondence methods seek on two given shapes for pairs of surface points that are semantically equivalent. We present three automatic algorithms that address three different aspects of this problem: 1) coarse, 2) dense, and 3) partial correspondence. In 1), after sampling evenly-spaced base vertices on shapes, we formulate the problem of shape correspondence as combinatorial optimization over the domain of all possible mappings of bases, which then reduces within a probabilistic framework to a log-likelihood maximization problem that we solve via EM (Expectation Maximization) algorithm. Due to computational limitations, we change this algorithm to a coarse-to-fine one (2) to achieve dense correspondence between all vertices. Our scale-invariant isometric distortion measure makes partial matching (3) possible as well.\n\n\n\nYusuf Sahillioglu\n\nKoç University, İstanbul/Turkey\n\nhttp://home.ku.edu.tr/~ysahillioglu/\n\n\n\n\n\n
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_672
DTSTAMP:20110718T072011
DTSTART:20110718T091500Z
DTEND:20110718T100000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGMB]: 3D Isometric Shape Correspondence
DESCRIPTION:\n\nSpeaker: Yusuf Sahillioglu - Koç University, İstanbul/Turkey\n\n3D Isometric Shape Correspondence\n\nAbstract:\n\n3D shape correspondence methods seek on two given shapes for pairs of surface points that are semantically equivalent. We present three automatic algorithms that address three different aspects of this problem: 1) coarse, 2) dense, and 3) partial correspondence. In 1), after sampling evenly-spaced base vertices on shapes, we formulate the problem of shape correspondence as combinatorial optimization over the domain of all possible mappings of bases, which then reduces within a probabilistic framework to a log-likelihood maximization problem that we solve via EM (Expectation Maximization) algorithm. Due to computational limitations, we change this algorithm to a coarse-to-fine one (2) to achieve dense correspondence between all vertices. Our scale-invariant isometric distortion measure makes partial matching (3) possible as well.\n\n\n\nYusuf Sahillioglu\n\nKoç University, İstanbul/Turkey\n\nhttp://home.ku.edu.tr/~ysahillioglu/\n\n\n\n\n\n
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Michael Black":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_673
DTSTAMP:20110718T072025
DTSTART:20110718T091500Z
DTEND:20110718T100000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGMB]: 3D Isometric Shape Correspondence
DESCRIPTION:\n\nSpeaker: Yusuf Sahillioglu - Koç University, İstanbul/Turkey\n\n3D Isometric Shape Correspondence\n\nAbstract:\n\n3D shape correspondence methods seek on two given shapes for pairs of surface points that are semantically equivalent. We present three automatic algorithms that address three different aspects of this problem: 1) coarse, 2) dense, and 3) partial correspondence. In 1), after sampling evenly-spaced base vertices on shapes, we formulate the problem of shape correspondence as combinatorial optimization over the domain of all possible mappings of bases, which then reduces within a probabilistic framework to a log-likelihood maximization problem that we solve via EM (Expectation Maximization) algorithm. Due to computational limitations, we change this algorithm to a coarse-to-fine one (2) to achieve dense correspondence between all vertices. Our scale-invariant isometric distortion measure makes partial matching (3) possible as well.\n\n\n\nYusuf Sahillioglu\n\nKoç University, İstanbul/Turkey\n\nhttp://home.ku.edu.tr/~ysahillioglu/\n\n\n\n\n\n
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Michael Black":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_677
DTSTAMP:20110808T091531
DTSTART:20110818T121500Z
DTEND:20110818T130000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGBU]: Attention-guided Augmentation of Animations and Stills 
DESCRIPTION:\n\nEakta Jain, Carnegie Mellon University, USA\n\nAttention-guided Augmentation of Animations and StillsArtists who work with visual media understand that human visual attention is a limited resource and train themselves to use this resource well. In this talk, I examine how viewer attention might influence algorithms that augment visual media with digitally created effects. I will present two projects that develop this central idea.In the first project, we leverage the techniques of 3D computer animation, such as physical simulation, for hand-drawn character animation. Researchers have attempted to compute accurate 3D reconstructions of the shape and motion of a hand-drawn character, which are meant to act as its `proxy' in the 3D environment. This task is challenging because in addition to the 2D-to-3D depth ambiguity, hand drawings also contain geometric inconsistencies. We propose that the 3D proxy for a hand-drawn character have different levels of detail, so that at each level the error terms account for quantities that might attract viewer attention. We show that the different levels of detail can be generated via three basic error terms.In the second project, we consider the problem of retargeting comic book artwork to digital display devices by the addition of 2D camera moves on the original still pictures.  We observe that artists compose their pictures to direct the viewer's attention along a deliberate route. By recording the gaze of viewers looking at the pictures, we aim to reconstruct the artist's intention in directing viewer attention. We then generate a camera move that respects the artist-intended visual route. We verify experimentally that the artist is indeed successful in directing viewer attention, and propose an algorithm to predict the parameters of a camera move on a still picture based on the recorded gaze data of multiple viewers looking at that picture.Short bio:Eakta Jain is a PhD candidate at Carnegie Mellon University, advisedby Jessica Hodgins and Yaser Sheikh. Eakta has interned at DisneyResearch, Pittsburgh and at the Walt Disney Animation Studios. Shereceived her Master of Science degree from Carnegie Mellon Universityin 2010 and the Bachelor of Technology degree from Indian Institute ofTechnology Kanpur in 2006. She received the Best Paper Award at SCA2010 for her work on augmenting hand-animation with physically basedsimulation. In her free time, she enjoys swimming, reading sciencefiction, sketching and watching animated films. Her personal webpage iswww.cs.cmu.edu/~ejain.Host: Betty Mohler
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Betty Mohler":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_712
DTSTAMP:20110921T091303
DTSTART:20110927T130000Z
DTEND:20110927T140000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGBU]: Integration of information in complex neural networks with noise
DESCRIPTION: Speaker: Jordi Grau Moya, Universitat,Politecnica de Catalunya, Barcelona, Spain\n\nTitle: Integration of information in complex neural networks with noise\n\n"Our aim in this talk is to study, by computer simulation, the capacity of the brain to transmit and process information in noisy environments, in particular, the dynamicalbehavior of a scale-free complex neuronal network. Specifically, we will focus on tonic neurons as the source of information, and study the effects of the topology on the propagationof the information to other neurons."
ORGANIZER;CN="Daniel Braun":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_722
DTSTAMP:20111011T083616
DTSTART:20111014T101500Z
DTEND:20111014T110000Z
CATEGORIES:DepKo
SUMMARY:[Depko-AGBU]: An investigation into the SIMONA motion base control algorithms
DESCRIPTION:Speaker: Guy Rigot, OHB SystemTitle: An investigation into the SIMONA motion base control algorithmsAbstract:The section of Control & Simulation at the Faculty of Aerospace Engineeringof Delft University of Technology operates a high fidelity, six degree offreedom flight simulator for ongoing research. At the time of writing ofthis thesis (2008), a full and thorough understanding of the implementedmotion base algorithms was not available. In order to enable futureimprovement of the algorithms, a thorough investigation into the motion basewas initiated.This thesis led to a thorough understanding (both from a theoretical andpractical point of view) of the implemented algorithms in the motion baseand their properties. Next to this understanding, a modular simulation modelof the motion base was created, allowing experimentation with other, moreadvanced control algorithms. Finally, a rudimentary identification of massand inertial properties was performed.
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_719
DTSTAMP:20110930T095021
DTSTART:20111103T101500Z
DTEND:20111103T110000Z
CATEGORIES:FriKo, DepKo
SUMMARY:[Friko]: The Disembodied Eye:  Consequences of Displacing Perception from Action
DESCRIPTION:Speaker: Roberta Klatzky, Carnegie Mellon University, USA\n\nTitle: The Disembodied Eye:  Consequences of Displacing Perception from Action\n\nAbstract:\n\nIn everyday life, when we perform a visually guided action, we expect to perceive in the same space where we act.  New technologies have developed that decouple perception from action, forcing people to look here and act there.  What are the consequences for perception?  For action? For spatial understanding?  I will describe research contrasting visual displays that are either co-located with the action they direct, or displaced to another location.  The consequences of displacing vision turn out to be quite broad, encompassing target localization, integrating vision and touch, performing visually directed action, motor learning, and mental visualization of wholes from parts. 
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Betty Mohler, AGBU":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_729
DTSTAMP:20111212T100743
DTSTART:20111213T101500Z
DTEND:20111213T110000Z
CATEGORIES:DepKo
SUMMARY:[Depko]: Errors, Expectations, Perception & Action
DESCRIPTION:Speaker: Gavin Buckingham\n\nTitle: Errors, Expectations, Perception & Action\n\nAbstract:\n\nIn most circumstances, individuals are able to make relatively accurate predictions about how heavy an object will be, even if they have never lifted it before. These expectations of how heavy an object is likely to be have dramatic effects on both how the object is lifted and how heavy it feels to the lifter. In the context of the famous ‘Size-Weight’ illusion, I will first outline the experiments showing how lifters’ expectations of heaviness can reliably and persistently affect our perception of how heavy an object feels. I will then describe how the error detection and correction processes which allow us to rapidly adapt our fingertip forces to an object’s veridical mass rely heavily upon vision. Finally, I will examine possible ways to utilize action observation and enhanced visual feedback to supplement normal motor learning processes in experimental object lifting tasks and applied minimally-invasive surgical training.
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Sally Linkenauger":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_726
DTSTAMP:20111123T162321
DTSTART:20120125T140000Z
DTEND:20120125T150000Z
CATEGORIES:DepKo
SUMMARY:[Depko]: The accuracy of Kinect segment tracking algorithms 
DESCRIPTION:Speaker: Simon Choppin & Jon Wheat, Sheffield Hallam University, UK\n\nTitle: The accuracy of Kinect segment tracking algorithms\n\nAbstract:\n\n Cameras which can ‘see’ into the third dimension have been available for many years, utilising structured light and time of flight techniques to calculate depth values. However, the release of Microsoft’s Kinect gaming accessory has heralded an explosion of interest in depth cameras due to its low cost and versatility. The Kinect captures colour video and depth information at 30 Hz and with the aid of proprietary algorithms is able to track objects and body segments.  Researchers working in robotics and computer vision have embraced this technology and have shown astonishing results with 3D scene reconstruction and robot navigation (amongst others). The Centre for Sports Engineering Research has followed developments in the Kinect with interest, particularly in the area of biomechanics and human motion analysis. Segment tracking with the Microsoft Kinect is a method of markerless human motion capture, made possible due to the real-time depth information provided by the device. Accurate tracking of human motion without markers is coveted by the biomechanics community due to the ease of setup and lack of physical interference. However, the accuracy of this tracking dictates which analytical applications are possible. In order to quantify the accuracy of Kinect segment tracking, a series of motions were recorded with the Kinect and a 12-camera Motion Analysis Corporation capture suite (MAC). Using the MAC system as a benchmark, Upper and lower body Euler angles, 3-point angles and planar segment angles were compared with Kinect data. Values calculated using the Kinect were obtained using both real-time and post-processing segment tracking algorithms. This presentation discusses the methods of markerless motion capture made possible with the Kinect, the accuracy of the systems given by our testing and the implications of this accuracy with regards to biomechanical analysis and computer animation.\n\n
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Betty Mohler, AGBU":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_781
DTSTAMP:20120208T165737
DTSTART:20120214T160000Z
DTEND:20120214T170000Z
CATEGORIES:DepKo
SUMMARY:[Depko]: Implicit Behavioral Contagion
DESCRIPTION:Title: Implicit Behavioral Contagion\n\nSpeaker: Katsumi Watanabe, Research Center for Advanced Science and Technology, The University of Tokyo\n\nAbstract:\n\nPeople tend to implicitly assimilate each other. In this talk, I present two studies on implicit behavioral contagion. The first sets of experiments concerns speed contagion, where people tended to modify their movement timing according to others' movements even when the observed and the to-be-executed movements are unrelated. Observers viewed either point-light biological motion, scrambled biological motion, or solid object motion. The stimulus sequence was presented at three different (half, normal, and double) rates. The observers performed a simple choice reaction-time task that was unrelated to the presented stimulus sequence. The observation of the biological motion produced a negative correlation between reaction time and stimulus speed, whereas no such trend was observed with the scrambled or solid object motion. These results suggest that behavioral tempo may be contagious; the speed of others' movements may automatically influence the timing of movement execution by the observer. In the second sets of experiments, we examined how the presence of another person and the history of interaction with the person would influence body movements by measuring involuntary body movements while people tried to keep their posture still. Participants were randomly paired and seated face-to-face. They were instructed to point each other with their index finger and to hold their fingers' absolute positions. Then, they performed an adaptation session where one participant moved his/her finger and arm freely and the other participant followed the partner's finger position as accurately as possible. After the adaptation session, they again asked to hold their finger positions still. The results showed that the peak cross-correlation between finger positions was significantly higher for the pairs that were actually paired than those made by random shuffling. Further, the adaptation increased the peak cross-correlation only when one participant intentionally pursued the position of the other participant’s finger. These results suggest that body movements tended to be mirrored implicitly and the short-term history of interpersonal interaction may affect this tendency. \n\n
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Tobias Meillinger":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_775
DTSTAMP:20120125T085946
DTSTART:20120326T091500Z
DTEND:20120326T100000Z
CATEGORIES:DepKo
SUMMARY:[Depko]: Regionally distinct processing of positive and negative valences by the primate ventromedial prefrontal cortex
DESCRIPTION:\n\nSpeaker: Dr. Ilya Monosov, National Eye Institute, USA\n\nTitle: Regionally distinct processing of positive and negative valences by the primate ventromedial prefrontal cortex\n\nAbstract: \n\nI will discuss the results of our recent neuronal recordings in the primateventromedial prefrontal cortex, an area that has been linked to dysfunctionsof mood such as major depression by clinical and fMRI studies in humans, butthat has not been studied at the level of single neurons. Our experimentsdemonstrate sub-regional differences within vmPFC in the processing of theprediction and the reception of rewards and punishments that may account forthe finding that continuous electrical stimulation of the human dorsal vmPFCrelieves symptoms of major depression (Mayberg, 2009). I will discuss theseresults and provide future directions for circuit-based analysis ofemotion-related circuits using multiple techniques, such as optogeneticmanipulation of neuronal activity.
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Conchy Moya":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_788
DTSTAMP:20120320T130253
DTSTART:20120329T090000Z
DTEND:20120329T100000Z
CATEGORIES:DepKo
SUMMARY:[Depko] "And if we simply fly to work?" – An early technology assessment of personal air transportation
DESCRIPTION:\n\n\n\n\n\nTitle: "And if we simply fly to work?" – An early technology assessment of personal air transportation.\n\nSpeakers: Michael Decker, Sarah Meyer\n\nInstitution:  Institute for Technology Assessment and Systems (ITAS), Karlsruhe Institute of Technology (KIT) \n\nAbstract:\n\nCommuter traffic causes peak loads on ground based traffic systems. Traffic jams in the morning and evening are indicators of this effect. The construction of streets and highways conflicts with other forms of land-use. Therefore, it is an interesting idea to use the third dimension – the airspace – in order to relax the commuter traffic situation on the ground. It combines the “old dream of a flying car” with an actual need for changes in the transportation system. A European FP7 project called “myCopter – Enabling Technologies for Personal Aerial Vehicles” is investigating the idea of a Personal Air Transportation System (PATS) with small vehicles that take-off and land vertically. Besides looking at desirable technical developments for such a PATS (like automation technologies for obstacles avoidance, path planning, etc.) the myCopter project consortium is especially interested in the associated challenges and obstacles which are connected with the realization of such a system from a technology assessment’s point of view.\n\nThe presentation will give an overview of the socio-technological implications surrounding the envisioned PATS with small personal air vehicles (PAVs) for the context of commuting in an urban environment. Important issues which were identified during the first year of the project are safety concerns (collision avoidance, controlled flight into terrain, terrorist threats, bad weather etc.), the absence of a clear legal framework for the certification and regulation of PAVs and steep training requirements that are currently enforced for the obtaining and maintaining a pilot license. Other questions are how to integrate the PAVs into the existing transportation systems and what the design of Air Traffic Management could look like. Connected with all the previous issues is the complex question of who is in authority over the pilot tasks and which level of automation should be considered.
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Frank Nieuwenhuizen":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_796
DTSTAMP:20120412T113521
DTSTART:20120510T130000Z
DTEND:20120510T140000Z
CATEGORIES:DepKo
SUMMARY:[Depko] Vision at the edge of the temporal field
DESCRIPTION:\n\nSpeaker: Michelle To, University of Hull, UK\n\nTitle: Vision at the edge of the temporal fieldThe temporal visual field extends to more than 100 degrees from the line of sight, and the far periphery is critical in detecting incoming threats and in monitoring optic flow for purposes of balance and locomotion.  Yet while detection of moving stimuli has been extensively studied in the central and near-peripheral regions of vision, it has rarely been investigated at extreme eccentricities.  In Experiment 1 we measured contrast sensitivity for stationary and moving gratings:  At all eccentricities from 70 deg outwards, sensitivity was higher for moving stimuli; and in the outermost region, only moving stimuli were visible.  Experiment 2 demonstrated a marked anisotropy for direction of drift in the extreme periphery:  Contrast sensitivity was highest close to the vertical and horizontal axes.  In addition, sensitivity was higher in the temporal to nasal direction than in the opposite direction.  Experiment 3 suggested that the anisotropy arises early in the visual pathway:  When plaid stimuli were compared with Gabors, the highest sensitivities occurred for drift directions intermediate to the cardinal directions, implying that the sensitivity was determined by the motions of the component gratings rather than those of the integrated plaid. This research is only a preliminary first step in investigating visual processing in the extreme periphery.\n\nHost: Betty Mohler
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Betty Mohler":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_815
DTSTAMP:20120510T152641
DTSTART:20120521T133000Z
DTEND:20120521T143000Z
CATEGORIES:DepKo
SUMMARY:[Depko] Recent advances in Neurotechnology
DESCRIPTION:Title: Recent advances in Neurotechnology\n\nSpeakers: Anne Porbadnigk, Stefan Haufe & Klaus Müller (TU Berlin, Germany)\n\nAbstract: The last years have seen a rise in interest in using BCI methodologyfor investigating non-medical questions beyond the purpose ofcommunication and control. The first part of the talk gives a shortintroduction to BCI from a machine learning perspective.  In theremaining 2 sections of the talk we present our work on selectedapplications of BCI.Part II discusses the use of EEG in combination with BCI methods forinvestigating how signal quality is processed (non-)consciously on aneural level (auditory/visual domain).In the third part of this talk we present results of an EEG study onforced emergency braking in a driving simulator. Participants had toclosely follow a computer-controlled lead vehicle, which would induceemergency situations by braking abruptely in randomintervals. Grand-average results (N=18) indicate that interpretableneural signatures as well as (to some extent) muscle activity at thelower leg allow for much earlier detection of the driver's intentionto perform emergency braking than behavioural measures, whileachieving the same detection accuracy.
LOCATION:MPI Lecture Hall
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_816
DTSTAMP:20120528T134052
DTSTART:20120531T113000Z
DTEND:20120531T121500Z
CATEGORIES:DepKo
SUMMARY:[Depko-Kyb] Ultra-fast Steady State Free Precession (SSFP) Imaging
DESCRIPTION:Title: Ultra-fast Steady State Free Precession (SSFP) Imaging\n\nSpeaker: Olivier Bieri, Universitaetsspital Basel\n\nAbstract: \n\nUltra-fast steady state free precession (SSFP) imaging refers to Cartesian SSFP imaging protocols using repetition times close to 1ms, as can be achieved with clinical whole body systems by fully exploiting and optimizing gradient switching patterns and RF pulse settings in combination with asymmetric echo readouts. Typically, an isotropic resolution of ~1.8mm can be achieved with repetition times TR ~ 1.4ms (TE ~ 0.5ms), providing high-resolution 3D artifact free balanced SSFP images even for targets with severe susceptibility variations, such as the lung.
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Tina Schroeder":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_817
DTSTAMP:20120528T134656
DTSTART:20120606T113000Z
DTEND:20120606T121500Z
CATEGORIES:DepKo
SUMMARY:[Depko-Kyb] New Concepts for Spatial Encoding for Ultrafast MRI. PatLoc and  MR-Encephalography
DESCRIPTION:Title: New Concepts for Spatial Encoding for Ultrafast MRI. PatLoc and  MR-Encephalography\n\nSpeaker: Prof. Juergen Henning, Universitaetsklinikum Freiburg\n\nAbstract: TBA
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Tina Schroeder":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_820
DTSTAMP:20120528T140638
DTSTART:20120629T121500Z
DTEND:20120629T130000Z
CATEGORIES:DepKo
SUMMARY:Simulation of Aircraft Crew Behaviour based on Cognitive Models
DESCRIPTION: Andreas Lüdtke, University of OldenbergTitle: Simulation of Aircraft Crew Behaviour based on Cognitive ModelsAbstract: The talk will present a cognitive aircraft crew model which is intended to be used as a tool to make psychological knowledge about human cognitive processes readily available to cockpit system designers to foster human-centred design. The cognitive model is based on the information processing paradigm of human cognition in the tradition of John R. Anderson and Allen Newell: the human brain gets input from the environment, matches this input to mental knowledge structures and produces output in form of motoric actions. Thereby the human information processor closes a continuous loop with the outer world. The focus of research according to this paradigm is on the type, structure and processing of information. The structures and processes are modelled in abstract "cognitive architectures". Our cognitive architecture has a modular structure that allows to integrate partial models from cognitive psychology encompassing aspects fo perception, decision making and motor performance. The model has been implemented in the programming language C and has been connected to a flight simulator. In this way it was possible to test the validity of the model by comparing its performance with the performance of human aircraft pilots. We investigated the interaction between pilots and a 4D Flight Management System. Bio:Andreas Lüdtke is a group leader for Human Centered Design at OFFIS (Institut für Informatik, Oldenburg). He was the coordinator for HUMAN (http://www.human.aero/), a FP7 project that developed methodologies to predict of human errors in the aircraft cockpit and, in doing so, inform the development of cockpit systems. 
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_828
DTSTAMP:20120710T073730
DTSTART:20120710T120000Z
DTEND:20120710T130000Z
CATEGORIES:DepKo
SUMMARY:[Depko] Reward, impulsivity, and control of movement vigor
DESCRIPTION:Title: Reward, impulsivity, and control of movement vigor\n\nSpeaker: Reza Shadmehr, John Hopkins University, USA\n\nAbstract: Healthy people differ in how they move: some like to move fast, while others like to move somewhat slower.  What determines the vigor of our movements?  Here, I focus on the control of eye movements and suggest that the purpose of any movement is to acquire a more rewarding state. I suggest that the way the brain discounts reward in time partly explains why individuals differ in how they move.
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Daniel Braun":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_844
DTSTAMP:20120730T083658
DTSTART:20120731T091500Z
DTEND:20120731T100000Z
CATEGORIES:DepKo
SUMMARY:[Depko] Motor Action Affects Memory, Not Perception
DESCRIPTION:Title: Motor Action Affects Memory, Not Perception\n\nSpeaker: Bruce Bridgeman (University of California, Santa Cruz, USA) \n\nAbstract: Perception facilitates two functions: learning about the environment and controlling real-time behavioral interactions with it. The relation between embodied perception and action can be studied in the context of memory, and the change detection that it makes possible. It is enhanced when there is a physical interaction between the observer and the visual stimulus. Perception is something you do, not something that happens to you. This is demonstrated in experiments detecting scene changes with or without the hands placed near the image, and in studies of motor action (throwing a ball into a hole) vs verbal information. We find that action changes memory, but leaves perception intact. 
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Markus Leyrer":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_868
DTSTAMP:20121112T091356
DTSTART:20121114T100000Z
DTEND:20121114T110000Z
CATEGORIES:DepKo
SUMMARY:Neuromodulation and motivation: Why dopamine isn't enough.
DESCRIPTION: Sebastien Bouret. ICM, Paris. Title: Neuromodulation and motivation: Why dopamine isn't enough. Abstract: Motivation involves multiple processes that regulate behaviors as a function of their expected outcome value. The evaluation of reward, in which dopamine plays a central role, is at the heart of motivational processes. There are, however, other key processes ranging from arousal to decision making, implying the involvement of other neuromodulators such as noradrenaline. We compare the activity of noradrenergic and dopaminergic neurons in monkeys performing behavioral tasks where motivational processes are manipulated systematically. These tasks enable a fine dissection of decision making and arousal as a function of internal and external determinants of the subjective outcome value. Using single unit recordings, we show that the activation of noradrenergic locus coeruleus neurons is associated with actions requiring a higher cost, potentially mediating the associated effort. This contrasts with the activity of dopaminergic neurons, generally activated in the more valuable 'low cost' conditions. Together, these data provide a new perspective on the relative implication of noradrenaline and dopamine in reward processing, decision making and vegetative arousal. This sheds a new light upon the contributions of these two major neuromodulatory systems in motivational processes that underly behavior. 
LOCATION:MRZ Seminar Room
ORGANIZER;CN="MPI for Biological Cybernetics, Lewis Chuang":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_891
DTSTAMP:20130114T151817
DTSTART:20130115T101500Z
DTEND:20130115T110000Z
CATEGORIES:DepKo
SUMMARY:Depko:Time-Space Diagrams: novel decision support for airtraffic controllers
DESCRIPTION:Title: Time-Space Diagrams: novel decision support for airtraffic controllers\n\nSpeaker: Prof. Max Mulder (TU Delft)\n\nAbout the speaker:\n\nMax is professor of the Control and Simulation department of the Faculty of Aerospace Engineering, where research is conducted into: Aircraft Modeling and Identification; Real Time Simulation and Software Engineering; Aircraft Control and Handling Qualities; Human-Machine Interfaces and Flight Deck Development; Avionics and Air Traffic Management; and Dynamics and Control of Space Vehicles.
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Frank Nieuwenhuizen":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_887
DTSTAMP:20130111T100640
DTSTART:20130121T140000Z
DTEND:20130121T150000Z
CATEGORIES:DepKo
SUMMARY:Depko: Interpersonal emotion regulation and the presentation of worry
DESCRIPTION:Brian Parkinson, Oxford University, UK\n\nTitle: Interpersonal emotion regulation and the presentation of worry\n\nThis talk considers interpersonal effects and functions of emotion presentation from a relation-alignment perspective. In particular, I focus on processes whereby one person’s emotions affect another’s, and on how people attempt to regulate other people’s emotions by regulating their own emotions or emotional expressions. These processes are illustrated using recent research into anxiety transfer in close personal relationships.Host: Ekaterina Volkova
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Ekaterina Volkova (AGBU)":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_905
DTSTAMP:20130205T141921
DTSTART:20130219T101500Z
DTEND:20130219T110000Z
CATEGORIES:DepKo
SUMMARY:Depko (AgBu): Cognitive and neural features of Congenital Prosopagnosia
DESCRIPTION:Speaker: Davide Rivolta (MPI for Brain Research, Frankfurt, Germany)\n\nTitle: Cognitive and neural features of Congenital Prosopagnosia\n\nAbstract:Faces are ubiquitous in our environment and are very important during human social interaction. Thus, it is not surprising that one of the domains that has received a great deal of research attention over the last few decades involves the investigation of the cognitive and neural aspects of face processing in humans. For most people, recognizing familiar faces is seemingly effortless. However, people who suffer from congenital prosopagnosia (CP) never develop this skill. Given its developmental nature and its often reported face-selectivity, CP represents a unique window into the understanding of the neuro-cognitive aspects of typical face processing. In the current talk I will focus on the cognitive and neural aspects of CP by describing the features of behavioural covert face recognition, by characterizing MEG results on the M170, and by reporting preliminary multivariate fMRI results.\n\nHost: Janina Esins
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Janina Esins":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_948
DTSTAMP:20130225T084614
DTSTART:20130226T101500Z
DTEND:20130226T110000Z
CATEGORIES:DepKo
SUMMARY:[Depko] Saccades, models, membranes and patients: from the clinic to the lab and back
DESCRIPTION:Speaker:Stefano Ramat, University of Pavia, Italy\n\nTitle: Saccades, models, membranes and patients: from the clinic to the lab and back\n\nAbstract.Saccades are rapid eye movements used to redirect the fovea from one object to another. They must be fast and accurate to support clear vision. They are the most frequent voluntary eye movement we perform, usually 2-3 per second. Human saccades follow a target jump within ~250 ms, are fast (up to ~600°/s), brief (typically ~30–100 ms) and although they are over so quickly that sensory information may not be used to control them online, they are accurate, and stop abruptly (i.e. with little subsequent ocular drift). The understanding of the saccadic system has driven scientific research for decades as it represents a great benchmark for testing hypotheses on motor control and adaptation, since it has all the characteristics of volitional motor systems while it benefits from the higher level of anatomical and physiological understanding that characterizes mechanisms depending on the brain stem rather than on the cortex. The talk will present how the most relevant contributions towards the understanding of the saccadic mechanism have been attained through the interaction between the clinical and the mathematical modeling approaches. It will then focus on the modeling of the mechanism responsible for saccadic oscillations showing an example of multiple scale modeling effort attempting to explain a behavioral finding in terms of both network topology and neuronal membrane properties.
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_964
DTSTAMP:20130415T125718
DTSTART:20130503T090000Z
DTEND:20130503T092500Z
CATEGORIES:DepKo
SUMMARY:[Depko] Telerobotics in offshore applications at Samsung Heavy Industries
DESCRIPTION:Speaker: Hyoung Il Son, Samsung Heavy Industries\n\nTitle: Telerobotics in offshore applications at Samsung Heavy Industries\n\nAbstract: In this talk, several projects progressed in Samsung Heavy Industries regarding (multi-user and multi-robot) teleoperation system for unmanned  offshore plants and unmanned underwater robots will be presented. The projects are undergoing from viewpoints of practical implementation rather than ones of theoretical contribution. Main goal of the projects is a commercialization of results (i.e. telerobot systems) so that this talk could give an insight to commercialize academical results in control and robotics fields.Bio: Hyoung Il Son received the B.S. and M.S. degrees from the School of Mechanical Engineering, Pusan National University, Pusan, Korea, in 1998 and 2000, respectively, and the Ph.D. degree from the Department of Mechanical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea, in 2010. He is now a principal research engineer and leader of Telerobotics Group at Institute of Industrial Technology, Samsung Heavy Industries, Daejeon, Korea. He was a Senior Researcher with LG Electronics (2003–2005) and Samsung Electronics (2005–2009) and a Research Associate with The University of Tokyo, Tokyo, Japan (2010). From 2010 to 2012, he was a Research Scientist with the Max Planck Institute for Biological Cybernetics, Tuebingen, Germany. His research interests include haptics, teleoperation, human-centered control design and evaluation, and supervisory control of discrete-event/hybrid systems. Host: Antonio Franchi
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Antonio Franchi (AgBu)":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_957
DTSTAMP:20130402T085942
DTSTART:20130503T093000Z
DTEND:20130503T095500Z
CATEGORIES:DepKo
SUMMARY:[Depko] Novel robotics based on the biarticular actuator coordinate
DESCRIPTION:Title: Novel robotics based on the biarticular actuator coordinate\n\nSpeaker: Dr. Sehooon Oh, Samsung Heavy Industries\n\nAbstract: \n\nRobotics has been developed based on the joint coordinates, where all the joints are considered disconnected from each other. However, humans and most of animals have the biarticular muscles in common which connect two joints and produce torques on both joints at the same time. When this biarticular muscle idea is applied to the robotics, various points of the robotics that are hard to find out in the conventional joint coordinates can be easily solved. This talk revisits the most fundamental points of robotics in terms of this biarticular muscle torque: 1) Statics, Kimematics and Dynamics are redefined using a novel coordinate using the biarticular muscle concept. 2) Advantages of the biarticular muscle structure are explored. 3) Implementation method of the biarticular muscle is explained. 4) The relationship between the biarticular muscle coordinate and the natural dynamics of walking is revealed. Lastly, conclusions and future challenges will be given.\n\nBio: Sehooon Oh received the B.S., M.S., and Ph.D. degrees in electrical engineering from the University of Tokyo, Tokyo, Japan, in 1998, 2000 and 2005 respectively. He worked as an assistant professor in the Department of Electrical Engineering and Information Systems of the University of Tokyo until 2012. Now he is a senior engineer at Samsung Heavy Industries. His research fields include the development of human friendly motion control algorithms a nd assistive devices for people. 
LOCATION:MRZ Seminar room
ORGANIZER;CN="Antonio Franchi, AGBU":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_973
DTSTAMP:20130502T084517
DTSTART:20130507T091500Z
DTEND:20130507T101500Z
CATEGORIES:DepKo
SUMMARY:[Depko] Visualizing Uncertainty
DESCRIPTION:Title: Visualizing Uncertainty\n\nSpeaker: William B. Thompson (School of Computing, University of Utah, USA)\n\nAbstract: Currently, most decisions made on the basis of computer models do not adequately account for the uncertainties inherent in those models.  This is because the tools for describing or depicting uncertainty in multidimensional data are not adequate for widespread use in real applications.  This talk will start by providing an overview of a large, multi-institutional project investigating the modeling, display, and understanding of uncertainty in simulations for policy decision making.  Next, results of an experiment probing the ability of people to make informed decisions when presented with visible indications of the intrinsic variability of the information on which such decisions are made will be described.  Finally, speculations will be presented about the role that uncertainty visualization should play in support of decision making.\n\nHost: Jeanine Stefanucci
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Jeanine Stefanucci":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_922
DTSTAMP:20130212T131248
DTSTART:20130513T091500Z
DTEND:20130513T100000Z
CATEGORIES:DepKo
SUMMARY:[Depko] An overview of the IDH team research on Human-inspired Cybernetic Systems
DESCRIPTION:scheduled May 14, 2013 from 11:15 AM to 12:15 PMLocation: MPH Lecture HallTITLE: An overview of the IDH team research on Human-inspired Cybernetic SystemsABSTRACTIn this talk, I will give an overview of the research activities carried out by the Interactive Digital Human (IDH) group. The group is based in part in Montpellier, France (LIRMM Laboratory) and in part in Tsukuba, Japan (CNRS/AIST Joint Robotics Laboratory). We focus on human-inspired cybernetic systems, mainly targeted at collaborative humanoid robotic applications. Four main scientific themes are studied at IDH: multimodal sensor-based (vision and force) control of humanoids, human/humanoid collaboration, human/humanoid embodiment (via BCI and fMRI) and human modeling for humanoid control. I will present the results obtained by our group in these fields within the European Projects VERE and RoboHow.Cog, and within the French Projects ANR ICARO and ANR ASSIST.BIO Andrea Cherubini is Associate Professor at Université de Montpellier 2 and Researcher at LIRMM IDH since 2011. He received an MSc in 2001 from the University of Rome « La Sapienza » and a second one in 2003 from the University of Sheffield, U.K. From 2004 to 2008, he was PhD student and then Postdoctoral fellow at the Dipartimento di Informatica e Sistemistica, University of Rome « La Sapienza », under the supervision of Prof. Giuseppe Oriolo. Then, from 2008 to 2011, he worked as PostDoc at INRIA Rennes with Dr. François Chaumette. He is currently involved, with IDH, in European projects VERE and RoboHow.Cog and in the French Project ANR ICARO. His main research interests include sensor-based control, humanoid robotics, and human-robot interaction. 
LOCATION:MPH Lecture Hall
ORGANIZER;CN="Antonio Franchi, AgBu":
END:VEVENT
BEGIN:VEVENT
UID:www.tuebingen.mpg.de_3_972
DTSTAMP:20130501T131815
DTSTART:20130607T120000Z
DTEND:20130607T130000Z
CATEGORIES:DepKo
SUMMARY:[Depko] TBA
DESCRIPTION:Speaker: Prof. Victoria Interrante (University of Minnesota, USA)\n\nHost: Betty Mohler
LOCATION:MRZ Seminar Room
ORGANIZER;CN="Betty Mohler":
END:VEVENT
END:VCALENDAR