Scene Flow Estimation On Argoverse 2
Métriques
EPE 3-Way
EPE Background Static
EPE Foreground Dynamic
EPE Foreground Static
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | EPE 3-Way | EPE Background Static | EPE Foreground Dynamic | EPE Foreground Static |
|---|---|---|---|---|
| deflow-decoder-of-scene-flow-network-in | 0.034295 | 0.004561 | 0.073231 | 0.025093 |
| fast-neural-scene-flow | 0.111820 | 0.090712 | 0.163388 | 0.081360 |
| zeroflow-fast-zero-label-scene-flow-via | 0.049392 | 0.013082 | 0.117688 | 0.017406 |
| seflow-a-self-supervised-scene-flow-method-in | 0.048590 | 0.005990 | 0.121375 | 0.018404 |
| scalable-scene-flow-from-point-clouds-in-the | 0.061960 | 0.004939 | 0.156392 | 0.024549 |
| neural-scene-flow-prior | 0.060606 | 0.034445 | 0.115796 | 0.031576 |
| Modèle 7 | 0.047316 | 0.002428 | 0.102975 | 0.036544 |