Seminar by Sabine Susstrunk
Opponent Color Revisited
Sabine Susstrunk
EPFL, Switzerland
Date: Monday, March 3rd, 2014
Time: 3:30 PM
Venue: CS101.Abstract:
According to the efficient coding hypothesis, the goal of the visual system should be to encode the information presented to the retina with as little redundancy as possible. From a signal processing point of view, the first step in removing redundancy is de-correlation, which removes the second order dependencies in the signal. This principle was explored in the context of trichromatic vision by Buchsbaum and Gottschalk, and later Ruderman et al., who found that linear de-correlation of the LMS cone responses matches the opponent color coding in the human visual system. And yet, there is comparatively few research in computer vision that explicitly model and incorporate color opponency into solving imaging tasks. A common perception is that “colors” are redundant and/or too correlated to be of any interest. We show with several applications, such as saliency and super-pixels, that considering opponent colors can significantly improve computer vision tasks. We have additionally extended the concept of "color opponency" to include near-infrared for applications such as scene recognition, object segmentation, and semantic image labeling.
About the speaker:
Sabine Susstrunk leads the Image and Visual Representation Group (IVRG) at the School of Information and Communication Sciences at Ecole Polytechnique Federale de Lausanne (EPFL) since 1999. Her research interests are in computational photography, multimedia, color image processing and computer vision, and image quality. She has authored and co-authored over 120 publications, of which 6 have received best paper/demo awards, and holds 7 patents. She received the IS&T/SPIE 2013 Electronic Imaging Scientist of the Year Award for her contributions to color imaging, computational photography, and image quality.