Scene Flow Estimation On Kitti 2015 Scene
평가 지표
Runtime (s)
D1-all
D2-all
Fl-all
SF-all
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | Runtime (s) | D1-all | D2-all | Fl-all | SF-all | Paper Title | Repository |
|---|---|---|---|---|---|---|---|
| EPC | 0.05 | 26.81 | 60.97 | 25.74 | (u003e60.97) | Every Pixel Counts: Unsupervised Geometry Learning with Holistic 3D Motion Understanding | - |
| EPC++ | 0.05 | 23.84 | 60.32 | 19.64 | (u003e60.32) | Every Pixel Counts ++: Joint Learning of Geometry and Motion with 3D Holistic Understanding | |
| Multi-Mono-SF | 0.063 | 27.33 | 30.44 | 18.92 | 39.82 | Self-Supervised Multi-Frame Monocular Scene Flow | |
| Self-Mono-SF | 0.09 | 31.25 | 34.86 | 23.49 | 47.05 | Self-Supervised Monocular Scene Flow Estimation |
0 of 4 row(s) selected.