Workshop

Nervous systems have evolved impressive abilities to extract useful information about the environment from images. Jumping spiders use their eight-eyed visual systems to detect prey, discriminate objects, and navigate;most mammals can readily segment moving objects in a scene and estimate velocities to mediate complex visuo-motor tasks; and primates can readily infer the 3D surface structure of their world from two dimensional images. How are these tasks - all of which lie beyond the current abilities of modern machine vision systems - accomplished by neural circuits in the brain? This week will bring together a combination of experimentalists and theorists who are attempting to understand the neural mechanisms of visual perception. This workshop is jointly sponsored by MSRI and the Redwood Neuroscience Institute. The Tuesday, February 8, 2005 session will be held at RNI. Because space at RNI is limited, attendees should apply in advance for Tuesday's session only by sending an email to Teri Fry tfry@rni.org . There will be a reception after the session at RNI, Tuesday, Feb. 8 at the Computer History Museum in Mountain View. Workshop Schedule Monday, February 7 9:00-10:00 Pam Reinagel (UC San Diego) “Coding in LGN and implications for V1” 10:00-10:15 Discussion 10:15-10:45 -- Morning Tea -- 6th floor 10:45-11:45 Jon Touryan (UC Berkeley) “Analysis of V1 complex cell receptive fields with complex stimuli” 11:45-12:00 Discussion 12:00- 2:00 -- Lunch -- 2:00- 3:00 Matteo Carandini (Smith-Kettlewell Eye Research Institute) “Receptive fields and suppressive fields in the early visual system” 3:00- 3:15 Discussion 3:15- 3:45 -- Afternoon Tea -- 6th floor 3:45- 4:45 Charles Gray (MSU Bozeman) “Multi-neuron response dynamics in cat V1 to the presentation of time-varying natural scenes” 4:45- 5:00 Discussion Tuesday, February 8 (RNI) 8:30- 9:30 Bus to RNI 9:45-10:45 Jack Gallant (UC Berkeley) “Neural Coding Beyond V1” 10:45-11:00 Discussion 11:00-11:15 -- Break -- 11:15-12:15 Greg DeAngelis (Washington University) “Roles of Area MT in Stereo Vision” 12:15-12:30 Discussion 12:30- 1:30 -- lunch -- 1:30- 2:15 Don Glaser (UC Berkeley) “The role of Cortical Noise and Stochastic Resonance in Some Illusory Motion Percepts” 2:15- 2:30 Discussion 2:30- 3:30 David Mumford (Brown University) -- Open discussion on the role of inhibitory neurons and methods for recording from large numbers of neurons. 3:30- 4:00 -- Break -- 4:00- 5:00 Jeff Hawkins (RNI) “How the Cortex Works” 5:00- 5:15 Discussion 6:00- 9:00 Reception/Dinner at the Computer History Museum, Mountain View 9:00- 10:00 Bus to Berkeley Wednesday, February 9 9:00-10:00 Mike Lewicki (Carnegie-Mellon University) “Density Component Models for Learning Hierarchical Structure in Natural Images” 10:00-10:15 Discussion 10:15-10:45 -- Morning Tea -- 6th floor 10:45-11:45 Terry Sejnowski (Salk Institute) "Multiplicative Mixers in Natural Images" 11:45-12:00 Discussion 12:00- 2:00 -- lunch -- 2:00- 3:00 Steven Zucker (Yale) “Frenet Geometry, Horizontal Connections, and Early Vision” 3:00- 3:15 Discussion 3:15-3:45 -- Afternoon Tea -- 6th floor 3:45- 4:45 Stu Geman (Brown University) “Invariance and Selectivity in the Ventral Visual Pathway” 4:45- 5:00 Discussion Thursday, February 10 9:00-10:00 Tai-Sing Lee (Carnegie-Mellon University) “Cortical mechanisms for visual inference” 10:00-10:15 Discussion 10:15-10:45 -- Morning Tea -- 6th floor 10:45-11:45 Antonio Torralba (MIT) “How scene context guides attention” 11:45-12:00 Discussion 12:00- 2:00 -- lunch -- 2:00- 2:45 Bin Yu (UC Berkeley) “A better statistical model for spike trains” 2:45- 3:00 Discussion 3:00- 3:45 Charles Anderson (Washington University) “Spatial-frequency tiling of V1 simple cells is predicted by signal-to-noise considerations.” 3:45- 4:00 Discussion 4:00- 4:30 -- Afternoon Tea -- 6th floor 4:30- 5:15 Bruno Olshausen (UC Davis/RNI) “Sparse coding and inference in visual cortex” 5:15- 5:30 Discussion Friday, February 11 Working groups and discussion