Zero Shot Action Recognition On Kinetics
評価指標
Top-1 Accuracy
Top-5 Accuracy
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
比較表
| モデル名 | Top-1 Accuracy | Top-5 Accuracy |
|---|---|---|
| ost-refining-text-knowledge-with-optimal | 75.1 | 94.6 |
| learning-a-deep-embedding-model-for-zero-shot | 23.6 | 49.5 |
| elaborative-rehearsal-for-zero-shot-action | 42.1 | 73.1 |
| alternating-gradient-descent-and-mixture-of | 76.8 | - |
| match-expand-and-improve-unsupervised | 71.6 | - |
| bidirectional-cross-modal-knowledge | 68.5 | 91.1 |
| label-embedding-for-image-classification | 23.4 | 50.3 |
| transferring-textual-knowledge-for-visual | 68.9 | 90.3 |
| video-text-modeling-with-zero-shot-transfer | 70.1 | 88.9 |
| expanding-language-image-pretrained-models | 65.2 | 86.1 |
| rethinking-zero-shot-action-recognition | 45.9 | 78.8 |
| locate-gat-modeling-multi-scale-local-context | 58.7 | - |
| elaborative-rehearsal-for-zero-shot-action | 37.1 | 69.3 |
| all-about-knowledge-graphs-for-actions | 22.3 | 49.7 |
| languagebind-extending-video-language | 64.1 | 85.7 |
| evaluation-of-output-embeddings-for-fine | 22.3 | 48.2 |
| devise-a-deep-visual-semantic-embedding-model | 23.8 | 51.0 |
| orthogonal-temporal-interpolation-for-zero | 70.6 | - |
| leveraging-temporal-contextualization-for | 78.1 | 95.7 |
| an-embarrassingly-simple-approach-to-zero | 22.9 | 48.3 |