Hierarchical Human Parsing on PASCAL Person-Part (test)
90.89Head AccuracyPGN
Evaluation results
| Method | Links | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| PGNBackbone=ResNet-1012022.03 | 90.89 | 75.12 | 55.83 | 64.61 | 55.42 | 41.57 | — | — | — | 95.33 | — | — | 68.4 | |
| HSSN DeepLabV3+Backbone=ResNet-1012022.03 | 90.19 | 78.72 | 70.67 | 69.71 | 61.15 | 60.44 | 95.86 | 71.56 | 98.2 | 97.18 | 97.69 | 88.2 | 75.44 | |
| BGNetBackbone=ResNet-1012022.03 | 90.18 | 77.44 | 68.93 | 67.15 | 60.79 | 59.27 | — | — | — | 97.12 | — | — | 74.42 | |
| PCNetBackbone=ResNet-1012022.03 | 90.04 | 76.89 | 69.11 | 68.4 | 60.78 | 60.14 | — | — | — | 96.78 | — | — | 74.59 | |
| HHPBackbone=ResNet-1012022.03 | 89.73 | 75.22 | 66.87 | 66.21 | 58.69 | 58.17 | 93.44 | 68.02 | 96.77 | 96.94 | 96.86 | 86.13 | 73.12 | |
| SemaTreeBackbone=ResNet-1012022.03 | 89.15 | 74.76 | 63.9 | 63.95 | 57.53 | 54.62 | 92.36 | 67.13 | 95.11 | 96.84 | 95.98 | 85.44 | 71.59 | |
| CNIFBackbone=ResNet-1012022.03 | 88.02 | 72.91 | 64.31 | 63.52 | 55.61 | 54.96 | 91.82 | 66.56 | 94.33 | 96.02 | 95.18 | 84.8 | 70.76 | |
| SPGNetBackbone=ResNet-1012022.03 | 87.67 | 71.41 | 61.69 | 60.35 | 52.62 | 48.8 | — | — | — | 95.98 | — | — | 68.36 | |
| DeepLabV3+Backbone=ResNet-1012022.03 | 87.02 | 72.02 | 60.37 | 57.36 | 53.54 | 48.52 | 90.07 | 65.88 | 93.02 | 96.07 | 94.55 | 84.01 | 67.84 |