[CVPR 2023] VideoMAE V2: Scaling Video Masked Autoencoders with Dual Masking

OpenMMLab's Next Generation Action Understanding Toolbox and Benchmark

[NeurIPS 2022 Spotlight] VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training

InternVideo: General Video Foundation Models via Generative and Discriminative Learning (https://arxiv.org/abs/2212.03191)

Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities

A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch

OpenMMLab Computer Vision Foundation

The official pytorch implementation of our paper "Is Space-Time Attention All You Need for Video Understanding?"

Setup a new machine without sudo!

Developed for the 2020 Google Winter Camp