Initial commit

2025-02-17 16:27:14 +08:00 · 2025-02-17 16:27:14 +08:00 · 2de3a40a8d
commit 2de3a40a8d
1 changed files with 3 additions and 0 deletions
--- a/README.md
+++ b/README.md
@ -0,0 +1,3 @@
+# videomae-base
+
+VideoMAE 模型在 Kinetics-400 数据集上以自监督方式进行了 1600 轮预训练。该模型由 Tong 等人在论文 VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training 中提出。