Computer Vision

Our experiments illustrate the merits of the proposed approach in challenging re-identification scenarios including crowded public spaces.

We propose a predictor that is based on a number of category specific features ( e.g., sample size, entropy, etc.) for whether independent or joint composite detector may be more accurate for a given conjunction.

We present an algorithmic framework for the early classification of human intentions, and use it to accurately predict future human motions when planning the path of a robot in an environment that is shared with humans.

We propose the notion of semi-supervised vocabulary-informed learning to alleviate the above mentioned challenges and address problems of supervised, zero-shot and open set recognition using a unified framework.

In this work we improve training of temporal deep models to better learn activity progression for activity detection and early detection tasks.