Human Object Interaction Detection On V Coco
Métriques
AP(S1)
AP(S2)
Time Per Frame(ms)
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | AP(S1) | AP(S2) | Time Per Frame(ms) |
|---|---|---|---|
| efficient-two-stage-detection-of-human-object | 60.7 | 66.2 | 64 |
| relational-context-learning-for-human-object | 68.8 | 71.0 | - |
| drg-dual-relation-graph-for-human-object | 51.0 | - | - |
| mining-cross-person-cues-for-body-part | 63.0 | 65.1 | - |
| polysemy-deciphering-network-for-robust-human | 53.34 | - | - |
| hoi-analysis-integrating-and-decomposing | 53.3 | 60.3 | - |
| qpic-query-based-pairwise-human-object | 58.3 | 60.7 | 63 |
| dirv-dense-interaction-region-voting-for-end | 56.1 | - | 68 |
| a-skeleton-aware-graph-convolutional-network | 53.1 | 57.9 | - |
| spatio-attentive-graphs-for-human-object | 54.2 | 60.9 | 500 |
| rlipv2-fast-scaling-of-relational-language | 72.1 | 74.1 | - |
| pose-based-modular-network-for-human-object | 51.8 | - | - |
| transferable-interactiveness-prior-for-human | 49.0 | - | 513 |
| ican-instance-centric-attention-network-for | 44.7 | - | - |
| glance-and-gaze-inferring-action-aware-points | 54.7 | - | - |
| transferable-interactiveness-knowledge-for | 49.1 | - | - |
| detecting-human-object-interactions-with-2 | 64.2 | 66.3 | 43 |
| hotr-end-to-end-human-object-interaction | 55.2 | 64.4 | - |
| efficient-two-stage-detection-of-human-object | 59.0 | 64.5 | 43 |
| consistency-learning-via-decoding-path | - | - | - |
| qpic-query-based-pairwise-human-object | 58.8 | 61.0 | 46 |
| learning-human-object-interactions-by-graph | 44.0 | - | - |
| vsgnet-spatial-attention-network-for | 51.76 | 57.0 | 312 |
| detecting-human-object-interactions-with-1 | 53.23 | - | - |
| mining-the-benefits-of-two-stage-and-one | 63.91 | 65.89 | - |
| detecting-human-object-interactions-with-2 | 65.3 | 67.1 | - |
| pastanet-toward-human-activity-knowledge | 51.0 | 57.5 | - |
| hoiclip-efficient-knowledge-transfer-for-hoi | 63.50 | 64.81 | - |
| consistency-learning-via-decoding-path | - | - | - |
| transferable-interactiveness-prior-for-human | 48.7 | - | - |
| exploring-structure-aware-transformer-over-1 | 66.0 | 70.7 | 74 |
| boosting-human-object-interaction-detection | 65.7 | 68.2 | - |
| efficient-two-stage-detection-of-human-object | 61.3 | 67.1 | 131 |
| ppdm-parallel-point-detection-and-matching | 51.1 | - | 71 |