DTA: Deformable Temporal Attention for Video Recognition.

Xiaohan Lei, Mengyu Yang, Gongli Xi, Yang Liu, Jiulin Li, Lanshan Zhang, Ye Tian

Browse the full IJCNN paper archive.