Holistic Features are Almost Sufficient for Text-to-Video Retrieval.

Kaibin Tian, Ruixiang Zhao, Zijie Xin, Bangxiang Lan, Xirong Li

Browse the full CVPR paper archive.