MGMAE: Motion Guided Masking for Video Masked Autoencoding.

Bingkun Huang, Zhiyu Zhao, Guozhen Zhang, Yu Qiao, Limin Wang

Browse the full ICCV paper archive.