Dense Pixel Correspondence Estimation On
Métriques
Viewpoint I AEPE
Viewpoint II AEPE
Viewpoint III AEPE
Viewpoint IV AEPE
Viewpoint V AEPE
Résultats
Résultats de performance de divers modèles sur ce benchmark
Nom du modèle | Viewpoint I AEPE | Viewpoint II AEPE | Viewpoint III AEPE | Viewpoint IV AEPE | Viewpoint V AEPE | Paper Title | Repository |
|---|---|---|---|---|---|---|---|
| SPyNet | 36.94 | 50.92 | 54.29 | 62.60 | 72.57 | Optical Flow Estimation using a Spatial Pyramid Network | |
| DGC-Net aff+tps+homo | 1.55 | 5.53 | 8.98 | 11.66 | 16.70 | DGC-Net: Dense Geometric Correspondence Network | |
| COTR +Interp. | 7.98 | - | - | - | - | COTR: Correspondence Transformer for Matching Across Images | |
| DeepMatching* | 5.84 | 4.63 | 12.43 | 12.17 | 22.55 | DeepMatching: Hierarchical Deformable Dense Matching | |
| RANSAC-DMP+ | 0.48 | 2.24 | 2.41 | 4.32 | 5.16 | Deep Matching Prior: Test-Time Optimization for Dense Correspondence | |
| PWC-Net | 4.43 | 11.44 | 15.47 | 20.17 | 28.30 | PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume | |
| COTR | 7.75 | - | - | - | - | COTR: Correspondence Transformer for Matching Across Images | |
| FlowNet2 | 5.99 | 15.55 | 17.09 | 22.13 | 30.68 | FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks |
0 of 8 row(s) selected.