יום שלישי, 6.6.2017, 11:30
חדר 1061, בניין מאייר, הפקולטה להנדסת חשמל
I will present our pipeline for optical flow computing, based on a CNN for generating local descriptors. I will focus on our recent research, where we show that the matching problem that underlies optical flow requires multiple strategies, depending on the amount of image motion and other factors. We then study the implications of this observation on training a deep neural network for representing image patches in the context of descriptor based optical flow. We propose a metric learning method, which selects suitable negative samples based on the nature of the true match. This type of training produces a network that displays multiple strategies depending on the input and leads to state of the art results on the KITTI 2012 and KITTI 2015 optical flow benchmarks.
Tal Schuster is currently finishing his MSc at the school of Computer Science in Tel Aviv University under the supervision of Prof. Lior Wolf. Starting September 17, Tal will be a PhD candidate in MIT, focusing on machine learning techniques for computer vision and natural language processing.