York’s 50th: Vision research conference looks at the world of 3-D

York’s Centre for Vision Research is investigating the world of 3-D at its upcoming International Conference on Vision in 3-D Environments, June 23 to 26 at York, as part of the University’s 50th anniversary celebrations.

The conference will feature talks from over 20 internationally recognized vision scientists with daily morning and afternoon sessions devoted to a particular contemporary issue on the topic of 3-D vision. It is expected to attract local and international faculty, graduate students and postdoctoral fellows in vision who wish to learn of the latest discoveries and advances in 3-D vision. All conference presentations will be held in Lecture Hall C in the basement of the Computer Science & Engineering Building, Keele campus.

On day one, three professors will discuss their research on visually guided motion: Marc Ernst, leader of the Independent Junior Research Group on Multi-sensory Perception and Action at the Max Planck Institute for Biological Cybernetics in Tübingen, Germany; Randy Flanagan, principal investigator of the Cognition & Action Laboratory at Queen’s University; and Pieter Medendorp of the Donders Institute for Brain, Cognition and Behaviour at Radboud University Nijmegen in the Netherlands. Medendorp will explore the complex changes that occur in the optic flow as people navigate through the environment.

On day two, stereopsis will be the topic of the morning discussion with internationally known scientists Suzanne McKee (left) of the Smith-Kettlewell Eye Research Institute and Brian Rogers of Oxford University discussing recent psychophysical research on stereoscopic depth perception and perspective. One of the leading researchers in the study of neural basis for stereopsis, Bruce Cumming of the United States National Institutes of Health, will discuss his electrophysiological studies of stereoscopic vision.

In the afternoon, William Warren (right), a professor of cognitive and linguistics sciences at Brown University, will look at issues of navigation along with Heinrich Bülthoff, director of the Max Planck Institute for Biological Cybernetics, and Professor Gregory Dudek, director of McGill University’s School of Computer Science. Dudek will discuss his recent computer vision experiments which focus on underwater exploration and navigation.

Cue integration and depth in naturalistic scenes will be explored on day three of the conference, featuring talks by scientists such as Edward Adelson of the Massachusetts Institute of Technology (MIT), Laurence Maloney of New York University along with Marty Banks of the University of California, Berkeley and Barbara Gillam of the University of New South Wales. Also, David Knill of the University of Rochester will talk about his influential work on the effects of context on cue combination.

Left: Heinrich Bülthoff

Day four of the conference provides a mix of speakers, including Ning Qian of Columbia University, who is an expert in computational modelling of human stereopsis and Andrew Parker of Oxford University, a leader in electrophysiological studies of stereopsis. In addition, Christopher Tyler of the Smith-Kettlewell Eye Research Institute in San Francisco will detail his work on 3-D spatial interactions.

“In the 2-D environment, visual stimuli tend to elicit either local interactions, in the form of a classical receptive field structure, or spatially distributed interactions, such as location-invariant suppression effects,” says Tyler. His work speaks to the complexity of neural encoding of 3-D space.

The final session of the conference will be devoted to natural scene perception with speakers who reflect a diverse set of approaches to the issue. This includes Aude Oliva (right) of MIT and Russell Epstein of the University of Pennsylvania, who will discuss their recent research on the perception of scene layout with Oliva focusing on behavioural approaches and Epstein on brain imaging.

Tai Sing Lee (left) of Carnegie Mellon University will take a different perspective and look at the role of natural scene statistics on 3-D surface perception later that afternoon. Images of natural scenes are highly complex and ambiguous because they are generated by a combination of many possible scene factors such as 3-D surface geometry, perspective, surface properties and lighting conditions, says Lee. Because of that he favours a statistical inference approach that seeks to model the joint probability between images and the underlying causes in natural scenes to the classical computer vision approach. Lee will discuss what he’s found during the conference.

There will be time for questions and discussion with the speakers during the sessions. In addition there will be a poster session which will include approximately 80 contributed presentations.

The collected set of talks presented at this conference will form the basis for an edited volume, titled Vision in 3-D Environments, edited by Laurence Harris, chair of York’s Department of Psychology and a professor in the Centre for Vision Research, and York computer science & engineering Professor Michael Jenkin. The book will be available at the conference.

For a complete list of speakers and the detailed conference program, visit the International Vision Conference Web site. For more information, contact Teresa Manini at manini@cvr.yorku.ca or one of the conference organizers, Laurie Wilcox at lwilcox@yorku.ca or Michael Jenkin at jenkin@cse.yorku.ca. For registration information, click here.