Semantic Correspondence on AP-10K Intra-species (test)
89.1PCK@0.10MARCO
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| MARCOSupervision=Supervised, DINOv2 Encoder=true, SD Encoder=false, Uses object masks at training=true2026.04 | 89.1 | 32.6 | 77.4 | |
| MARCOSupervision=Supervised, DINOv2 Encoder=true, SD Encoder=false2026.04 | 88.7 | 31.1 | 77 | |
| SimpleMatchInput size=448x4482026.01 | 88.2 | 30.1 | 75.7 | |
| Ours (Adapt. Pose)†Supervision=Supervised, Adaptive Pose Alignment=true, Index Flipping=true2023.11 | 87.7 | 23.2 | 73.2 | |
| GeoAware-SC2026.01 | 87.7 | 23.2 | 73.2 | |
| Geo-SCSupervision=Supervised, DINOv2 Encoder=true, SD Encoder=true, Uses object masks at inference=true2026.04 | 87.7 | 23.2 | 73.2 | |
| OursSupervision=Supervised2023.11 | 87.5 | 23.1 | 73 | |
| GECOSupervision=Supervised, DINOv2 Encoder=false, SD Encoder=true, Uses object masks at training=true2026.04 | 82.5 | 19.2 | 67.1 | |
| SD+DINO (S)Supervision=Supervised2023.11 | 77 | 9.9 | 57 | |
| SD+DINO(S)2026.01 | 77 | 9.9 | 57 | |
| SD+DINOv2Supervision=Supervised, DINOv2 Encoder=true, SD Encoder=true2026.04 | 77 | 9.9 | 57 | |
| SD+DINO (S)Supervision=Supervised2026.07 | 77 | 9.9 | 57 | |
| DIY-SCSupervision=Unsupervised / Weakly Supervised, DINOv2 Encoder=true, SD Encoder=true, Uses object masks at training=true2026.04 | 70.6 | — | — | |
| Ours-Zero-Shot+Supervision=Zero-shot2023.11 | 68.7 | 11.3 | 49.8 | |
| MREPSupervision=Unsupervised2026.07 | 63 | 7.8 | 50.4 | |
| SD+DINOSupervision=Zero-shot2023.11 | 62.9 | 7.6 | 43.5 | |
| SD+DINOv2Supervision=Unsupervised / Weakly Supervised, DINOv2 Encoder=true, SD Encoder=true2026.04 | 62.9 | 7.6 | 43.5 | |
| DHFSupervision=Supervised2023.11 | 62.7 | 8 | 45.8 | |
| DHF2026.01 | 62.7 | 8 | 45.8 | |
| DHFSupervision=Supervised, DINOv2 Encoder=false, SD Encoder=true2026.04 | 62.7 | 8 | 45.8 | |
| DHFSupervision=Supervised2026.07 | 62.7 | 8 | 45.8 | |
| DiTFSupervision=Unsupervised2026.07 | 62.2 | 7.6 | 49.7 | |
| DINOv2+NNSupervision=Zero-shot2023.11 | 60.9 | 6.4 | 41 | |
| DINOv2+NNSupervision=Unsupervised / Weakly Supervised, DINOv2 Encoder=true, SD Encoder=false2026.04 | 60.9 | 6.4 | 41 | |
| DINOv2+NNSupervision=Unsupervised2026.07 | 60.9 | 6.4 | 41 | |
| DIFTSupervision=Zero-shot2023.11 | 50.3 | 6.2 | 34.8 | |
| DIFTSupervision=Unsupervised / Weakly Supervised, DINOv2 Encoder=false, SD Encoder=true2026.04 | 50.3 | 6.2 | 34.8 | |
| DIFTSupervision=Unsupervised2026.07 | 50.3 | 6.2 | 34.8 | |
| w/o MREPSupervision=Unsupervised2026.07 | 32.2 | 3.5 | 22.4 |