There has been significant progress in Masked Image Modeling (MIM). Existing MIM
methods can be broadly categorized into two groups based on the reconstruction target …

Masked Image Modeling (MIM) has achieved promising progress with the advent of Masked
Autoencoders (MAE) and BEiT. However, subsequent works have complicated the …

This paper focuses on self-supervised video representation learning. Most existing
approaches follow the contrastive learning pipeline to construct positive and negative pairs …