Video Classification On Youtube 8M
評価指標
Hit@1
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
モデル名 | Hit@1 | Paper Title | Repository |
|---|---|---|---|
| DCGN (self-attention graph pooling) | 87.7 | Hierarchical Video Frame Sequence Representation with Deep Convolutional Graph Network | - |
| Hierarchical LSTM with MoE | 86.8 | Efficient Video Classification Using Fewer Frames | |
| Mixture-of-2-Experts | 70.1 | YouTube-8M: A Large-Scale Video Classification Benchmark |
0 of 3 row(s) selected.