Computational Neuroscience: Vision

A Summer Course at Cold Spring Harbor Laboratories
June 28 - July 11, 1996

Summer 1998 course

Organizers:

David J. Heeger, Stanford University
Eero P. Simoncelli University of Pennsylvania
Michael N. Shadlen, University of Washington

Description:

Computational approaches to neuroscience have produced important advances in our understanding of neural processing. Prominent successes have come in areas where strong inputs from neurobiological, behavioral and computational approaches can interact. Through a combination of lectures and hands-on experience with a computer laboratory, this intensive course will examine several areas, including feature extraction, motion analysis, binocular stereopsis, color vision, higher level visual processing, visual neural networks, and oculomotor function. The theme is that an understanding of the computational problems, the constraints on solutions to these problems, and the range of possible solutions can help guide research in neuroscience. Students should have experience in neurobiological or computational approaches to visual processing. Some background in mathematics will be beneficial.

Course Organization and Format:

The course will follow the general format of the past four years that it was held. It will be two weeks long, and each day of the course will include both lecture/discussion periods and time on the computers. In past years, participants' course activities have run from 9 am through midnight. A typical day of the course involves two lectures and two formal computer laboratories, combined with periods of free discussion.

Computer Labs and Course Projects:

The computer labs will consist mainly of a series of computer tutorials. Some of these will cover the background material (linear systems theory, signal/image processing) that form the theoretical basis for much of the work on computational vision. But most of the tutorials will correspond to each of the lecture topics (see below). This year, MATLAB, will be used for most of the computer labs. As in past years, the participants will also be encouraged to do a course project, implementing a computational model of some aspect of vision.

Lecturers:

This year's lecturers will be: Ted Adelson (MIT), David Brainard (UC Santa Barbara), Dennis Dacey (University of Washington), Paul Glimcher (NYU), Norma Graham (Department of Psychology), John Maunsell (Baylor College of Medicine), Suzanne McKee (Smith-Kettlewell Eye Research), Fred Miles (NIH), Tony Movshon (NYU), John Palmer (University of Washington), Clay Reid (Harvard), Brian Wandell (Stanford).

Preliminary schedule:

June 28 am: Introduction
June 28 pm: Elements of early vision (Adelson)
June 29 am: Color (Brainard)
June 29 pm: Color constancy (Brainard) and lightness/brightness (Adelson)
June 30 am: Retina (Dacey)
June 30 pm: Retina (Dacey) and student presentations
July 1 am: Functional brain imaging (Wandell)
July 1 pm: Geniculo-cortical pathway (Reid)
July 2 am: V1 physiology (Movshon)
July 2 pm: V1 model (Heeger)
July 1 am: V1 physiology (Movshon)
July 3 am: Light adaptation, pattern detection and masking (Graham, Heeger)
July 3 pm: Texture (Graham)
July 4: day off
July 5 am: MT physiology (Shadlen)
July 5 pm: MT model (Simoncelli)
July 6: Motion psychophysics and models (Shadlen, Simoncelli)
July 7 am: Stereo (McKee)
July 7 pm: afternoon off
July 8 am: Pursuit (Miles)
July 8 pm: Visual Stabilization (Miles)
July 9 am: Saccades (Glimcher)
July 9 pm: Attention psychophysics (Palmer)
July 10: Attention physiology (Maunsell)
July 11 am: Course project presentations