Ivan LAPTEV (Willow) : Weakly supervised learning from images and video
Recent progress in visual recognition goes hand-in-hand with the supervised learning and large-scale training data.
While the amount of existing images and videos is huge, their detailed annotation is expensive and often ambiguous. To address these problems, we focus on the weakly-supervised learning using incomplete and noisy supervision for training. We will discuss recognition in still images using weakly-supervised convolutional networks for recognizing objects and human actions. We will also discuss the learning of human actions from videos and narrations. This talk will give a short overview in visual recognition and will outline future challenges and opportunities.