6th Annual Perceptual Science Forum

Recognition in computer vision is beginning to work -- one of the hot current lines of inquiry is what we should be recognizing. Recent work -- our own and others -- has explored increasing the "label space" for recognition toward large numbers of semantic labels embedded in a hierarchy, toward multiple attribute labels, and toward detailed spatial parsing. Predictions of these labels are improving results on problems from face recognition to large scale similar image retrieval. At the same time there are unavoidable challenges in large scale computation that must be met. I will present some of our results in each of these directions and try to motivate some of the wide open problems in this area.

The poster session followed shortly afterwards and included the following posters:

IGERT in Interdisciplinary Perceptual Science, http://perceptualscience.rutgers.edu

 

Below are some photos of the poster session:

Simple Image Gallery Extended