Research code for CVPR 2022 paper "SwinBERT: End-to-End Transformers with Sparse Attention for Video Captioning"
By chatting or signing in you agree to the Terms and chat-message logging (revocable in History).