BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:A fresh look on Vision
DTSTART:20141110T110000
DTEND:20141110T120000
DTSTAMP:20260407T025803Z
UID:42fa46cf038fa942311a097372f34297f5107d6fcf93cc9fbc3a9470
CATEGORIES:Conferences - Seminars
DESCRIPTION:Dr Michael Herzog\, EPFL SV BMI LPSY\nTextbooks on vision expl
 ain that information processing proceeds in a hierarchical and feedforward
  manner. After photo-transduction in the retina\, features of an object ar
 e analyzed by neurons in primary visual cortex V1 that are sensitive to li
 nes and edges. These neurons serve as the input to neurons in higher visua
 l areas that code for more complex features\, such as shapes. For example\
 , a neuron in V4\, coding for a square\, is modelled by two V1 neurons cod
 ing for vertical and two V1 neurons coding for horizontal lines\, that all
  project to the V4 neuron. In this perspective\, information processing pr
 oceeds from low level to high level feature analysis\, well in line with e
 xperimental and clinical results. Such hierarchical\, feedforward models a
 re not only the basis for most vision models in neuroscience but also for 
 engineering and computer science. However\, this perspective struggles to 
 explain why state of the art vision systems are still unable to find a pai
 r of scissors on a cluttered desk\, which is an easy exercise for humans. 
 With various examples\, I will show experimentally and conceptually that t
 he textbook model misses an important step in the analysis of object recog
 nition: flexible Gestalt based figure-ground segmentation. In this alterna
 tive perspective\, high level processing determines low level processing a
 s much as the other way around.
LOCATION:SV 1717A http://plan.epfl.ch/?lang=en&room=sv+1717a
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR