1 paper · 1 filter
Ayush Gupta, Anirban Roy, Rama Chellappa +3
We address the problem of video question answering (video QA) with temporal grounding in a weakly supervised setup, without any temporal annotations. Given a video and a question,…