[NeurIPS 2022 Spotlight] VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training
项目介绍
[NeurIPS 2022 精选论文] VideoMAE：遮挡自动编码器是自监督视频预训练中的数据高效学习器【此简介由AI生成】
Others Python64提交数action-recognitionmaemasked-autoencoderneurips-2022pytorchself-supervised-learningtransformervideo-analysisvideo-representation-learningvideo-transformervideo-understandingvision-transformer
定制我的领域