Action Recognition In Videos On Activitynet
評価指標
mAP
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 | mAP |
|---|---|
| 2d-or-not-2d-adaptive-3d-convolution | 84.0 |
| do-less-and-achieve-more-training-cnns-for | 53.8 |
| transferring-textual-knowledge-for-visual | 96.9 |
| internvideo2-scaling-video-foundation-models | 95.9 |
| do-less-and-achieve-more-training-cnns-for | 52.3 |
| smart-frame-selection-for-action-recognition | 84.4 |
| towards-universal-representation-for-unseen | 53.8 |
| learning-spatio-temporal-representation-with | 78.9 |
| fine-grained-video-categorization-with | 83.4 |
| multi-agent-reinforcement-learning-based | 90.05 |
| dynamic-sampling-networks-for-efficient | 87.9 |
| nsnet-non-saliency-suppression-sampler-for | 94.3 |
| listen-to-look-action-recognition-by | 89.9 |
| dsanet-dynamic-segment-aggregation-network | 90.5 |
| bidirectional-cross-modal-knowledge | 96.1 |
| temporal-saliency-query-network-for-efficient | 93.7 |