Hosted by the Generalisation in Mind & Machine research groug

• Join via Zoom: https://bristol-ac-uk.zoom.us/j/99214845591?pwd=SXU5QkVKcHArRCtSUHFjUmtIcVRQQT09
• Meeting ID: 992 1484 5591
• Passcode: 149507

​The talk will start with describing how symmetry is used in Mathematics and Physics. This introduction will conclude with a statement that symmetry plays an absolutely fundamental role in both of these sciences. I will then illustrate how symmetry of objects provides the basis for solving Figure-Ground Organization in vision. This will introduce the concept of veridicality in 3D vision. This concept will be extended to the case of viewing 3D shapes from 2D pictures. A set of examples of symmetry in natural objects will be followed by showing how symmetry is used as an a priori constraint (aka prior) in solving the ill-posed inverse problem of 3D shape recovery. A psychophysical experiment will be described and examples of recovery of shapes of real objects will be provided.​

Short Bio:  Zyg Pizlo received his MSc and PhD degrees in Electronic Engineering in Warsaw, Poland in 1978 and 1982, respectively. He then started studying human visual perception and problem solving. He received his PhD degree in Psychology from the University of Maryland, College Park in 1991. He spent 26 years at Purdue University and then moved to the University of California, Irvine. He is an author of 3 books on shape and one book on problem solving. His research has always revolved around solving ill-posed inverse problems in perception and cognition. These problems have traditionally been considered insoluble by psychologists and engineers, but humans solve them instantly and nearly perfectly.