VideoMAE: MaskedAutoencodersareData-Efficient LearnersforSelf-SupervisedVideoPre-Training

Open in new window