Actor Semantic Segmentation on A2D (test)
73.7Class-Avg Pixel AccKalogeiton et al. [10]
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Kalogeiton et al. [10]2018.03 | 73.7 | 90.6 | 49.5 | |
| fully-convolutional modelInput type=actor and action pairs as text2018.03 | 71.4 | 92.8 | 53.7 | |
| Xu et al. [28]2018.03 | 58.3 | 85.2 | 33.4 | |
| fully-convolutional modelInput type=actor and action pairs as text2018.03 | 52.4 | 91.7 | 34.8 | |
| Kalogeiton et al. [10]2018.03 | 47.5 | 88.7 | 29.7 | |
| Xu et al. [29]2018.03 | 45.7 | 74.6 | — | |
| Xu et al. [28]2018.03 | 43.3 | 84.2 | 19.9 | |
| Xu et al. [29]2018.03 | 25.4 | 76.2 | — |