"Inference for Vision"
Bill Freeman, MERL
Friday, April 20, 2001 at 2:00 P.M.
Lubrano Conference Room
We describe a learning-based method for low-level vision problems--estimating scenes from images. We generate a synthetic world of scenes and their corresponding rendered images, modeling their relationships with a Markov network. Given image data, we seek to infer the most probable scene explanation. Building on recent theoretical and experimental results, we use the belief propagation update rules, which are exact only for networks without loops, even on our Makov network with many loops. This allows us to efficiently find a local maximum of the posterior probability for the scene, given an image. We call this approach VISTA--Vision by Image/Scene Training.
We apply VISTA to the ``super-resolution'' problem (estimating high frequency details from a low-resolution image), showing state-of-the-art results. To illustrate the potential breadth of the technique, we also apply it for the motion estimation problem in a ``blobs world.'' We show figure/ground discrimination, solution of the aperture problem, and filling-in arising from application of the same probabilistic machinery.
Note: The speaker will be available to answer questions about the company.
Host: Michael Black