Pose Estimation on SYMSOL I (test)
5.11Avg Log Likelihood (avg)Ours (Induction Layer)
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Ours (Induction Layer)Training=Single model on all classes, Backbone=ResNet2023.07 | 5.11 | 4.91 | 4.22 | 6.1 | 5.73 | 4.69 | |
| Murphy et al.Training=Single model on all classes, Backbone=ResNet2023.07 | 4.1 | 4.45 | 4.26 | 5.7 | 4.81 | 1.28 | |
| Klee et al.Training=Single model on all classes, Backbone=ResNet2023.07 | 3.41 | 3.75 | 3.1 | 4.78 | 3.27 | 2.15 | |
| Gilitschenski et al.Training=Single model on all classes, Backbone=ResNet2023.07 | -0.43 | 3.84 | 0.88 | -2.29 | -2.29 | -2.29 | |
| Deng et al.Training=Single model on all classes, Backbone=ResNet2023.07 | -1.48 | 0.16 | -0.95 | 0.27 | -4.44 | -2.45 | |
| Prokudin et al.Training=Single model on all classes, Backbone=ResNet2023.07 | -1.87 | -3.34 | -1.28 | -1.86 | -0.5 | -2.39 |