Zero Shot Action Recognition On Ucf101
評価指標
Top-1 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 |
|---|---|
| semantic-embedding-space-for-zero-shot-action | 10.9 |
| cross-modal-representation-learning-for-zero | 58.7 |
| evaluation-of-output-embeddings-for-fine | 12.0 |
| match-expand-and-improve-unsupervised | 78.2 |
| leveraging-temporal-contextualization-for | 85.4 |
| bidirectional-cross-modal-knowledge | 86.6 |
| multimodal-open-vocabulary-video | 82.6 |
| objects2action-classifying-and-localizing | 30.3 |
| expanding-language-image-pretrained-models | 72.0 |
| zero-shot-action-recognition-with-error | 15.1 |
| transferring-textual-knowledge-for-visual | 85.8 |
| victr-video-conditioned-text-representations | 72.4 |
| locate-gat-modeling-multi-scale-local-context | 76.0 |
| an-embarrassingly-simple-approach-to-zero | 15.0 |
| モデル 15 | 14.9 |
| video-text-modeling-with-zero-shot-transfer | 86.6 |
| alignment-uniformity-aware-representation | 58 |
| claster-clustering-with-reinforcement | 53.9 |
| eva-clip-improved-training-techniques-for | 83.1 |
| i-know-the-relationships-zero-shot-action | 34.2 |
| orthogonal-temporal-interpolation-for-zero | 92.8 |
| multimodal-open-vocabulary-video | 87.1 |
| rethinking-zero-shot-video-classification-end | 48 |
| synthetic-sample-selection-for-generalized | 40.9 |
| モデル 25 | 15.9 |
| alternating-gradient-descent-and-mixture-of | 91.5 |
| alternative-semantic-representations-for-zero | 24.4 |
| multi-task-zero-shot-action-recognition-with | 15.8 |
| elaborative-rehearsal-for-zero-shot-action | 51.8 |
| モデル 30 | 16.7 |
| ez-clip-efficient-zeroshot-video-action | 79.1 |
| rethinking-zero-shot-action-recognition | 56.0 |
| evaluation-of-output-embeddings-for-fine | 9.9 |
| towards-universal-representation-for-unseen | 17.5 |
| ost-refining-text-knowledge-with-optimal | 79.7 |