Semantic Correspondence on SPair-71k (test)
88.9PCK@0.1SEMALIGN3D
Evaluation results
| Method | Links | ||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| SEMALIGN3D2025.03 | 88.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 77.5 | 15.8 | 80.5 | 72.5 | 15.4 | 76.9 | 67.3 | 46.6 | — | — | — | — | — | — | |
| SemAlign3DSupervision Type=Supervised2026.05 | 88.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 77.5 | 15.8 | — | — | — | — | — | — | — | — | — | — | — | — | |
| SimpleMatchBackbone=DINOv2, Resolution=448 x 4482026.01 | 88.2 | — | 79.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 26 | 3 | — | — | — | — | |
| MARCOSupervision Type=Supervised2026.05 | 87.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 77.6 | 27 | — | — | — | — | — | — | — | — | — | — | — | — | |
| GeoAware2025.03 | 85.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 75.3 | 22 | 80.7 | 73.2 | 22.7 | 75.3 | 66.3 | 48.9 | — | — | — | — | — | — | |
| SimpleMatchBackbone=DINOv2, Resolution=252 x 2522026.01 | 84.2 | — | 69.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 65 | 2.8 | — | — | — | — | |
| GeoAware-SCBackbone=SD+DINOv2, Resolution=960 x 8402026.01 | 83.2 | — | 72.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 11.8 | — | — | — | — | |
| Geo-SCSupervision Type=Supervised2026.05 | 83.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 72.8 | 21.7 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Jamais VuSupervision Type=Supervised2026.05 | 82.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 71.9 | 20.5 | — | — | — | — | — | — | — | — | — | — | — | — | |
| SD4MatchBackbone=SD, Resolution=768 x 7682026.01 | 75.5 | — | 59.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 1 | 12.5 | — | — | — | — | |
| Fuse-ViT-B/14supervision=Strong Supervised (S), protocol=per image PCK, trained bottleneck layer=true2023.05 | 74.6 | — | — | 81.2 | 66.9 | 91.6 | 61.4 | 57.4 | 85.3 | 83.1 | 90.8 | 54.5 | 88.5 | 75.1 | 80.2 | 71.9 | 77.9 | 60.7 | 68.9 | 92.4 | 65.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SD+DINOv2Supervision Type=Supervised2026.05 | 74.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 57.7 | 9.6 | — | — | — | — | — | — | — | — | — | — | — | — | |
| GECOSupervision Type=Supervised2026.05 | 73.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 59.6 | 14.2 | — | — | — | — | — | — | — | — | — | — | — | — | |
| 3D-SCSupervision Type=Weakly Supervised without human annotations2026.05 | 73 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 54.8 | 10.2 | — | — | — | — | — | — | — | — | — | — | — | — | |
| 3D-SCSupervision level=Weakly supervised without human annotations2026.05 | 73 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 54.8 | 10.2 | — | — | — | — | — | — | — | — | — | — | — | — | |
| SimpleMatchBackbone=iBOT, Resolution=256 x 2562026.01 | 72.2 | — | 55.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 73 | 3.1 | — | — | — | — | |
| DIY-SCSupervision Type=Weakly Supervised with human annotations2026.05 | 71.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 53.8 | 10.1 | — | — | — | — | — | — | — | — | — | — | — | — | |
| DIY-SCSupervision level=Weakly supervised with human annotations2026.05 | 71.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 53.8 | 10.1 | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINOv2-ViT-B/14supervision=Strong Supervised (S), protocol=per image PCK, trained bottleneck layer=true2023.05 | 69.9 | — | — | 80.4 | 60.2 | 88.1 | 59.5 | 54.9 | 82 | 73.5 | 89.1 | 53.3 | 85.5 | 73.6 | 73.8 | 65.2 | 72.3 | 43.6 | 65.6 | 91.4 | 60.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DIY-SC+OriAnySupervision Type=Weakly Supervised without human annotations2026.05 | 69.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 51.2 | 9.5 | — | — | — | — | — | — | — | — | — | — | — | — | |
| DIY-SC+OriAnySupervision level=Weakly supervised without human annotations2026.05 | 69.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 51.2 | 9.5 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Shape-of-You2026.03 | 67.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 50.8 | 10 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Stable Diffusionsupervision=Strong Supervised (S), protocol=per image PCK2023.05 | 66.3 | — | — | 75.6 | 60.3 | 87.3 | 41.5 | 50.8 | 68.4 | 77.2 | 81.4 | 44.3 | 79.4 | 62.8 | 67.7 | 64.9 | 71.6 | 57.8 | 53.3 | 89.2 | 65.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LPMFlowBackbone=iBOT, Resolution=256 x 2562026.01 | 65.6 | — | 46.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 36 | 5.6 | — | — | — | — | |
| DHFBackbone=SD, Resolution=512 x 5122026.01 | 64.9 | — | 50.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 12.5 | — | — | — | — | |
| DHFSupervision Type=Supervised2026.05 | 64.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 50.2 | 8.7 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Spherical Map.Supervision Type=Weakly Supervised with human annotations2026.05 | 64.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 48.2 | 8.4 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Spherical Map.Supervision level=Weakly supervised with human annotations2026.05 | 64.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 48.2 | 8.4 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Fuse-ViT-B/14supervision=Unsupervised nearest neighboring (UN), protocol=per point PCK2023.05 | 64 | — | — | 73 | 64.1 | 86.4 | 40.7 | 52.9 | 55 | 53.8 | 78.6 | 45.5 | 77.3 | 64.7 | 69.7 | 63.3 | 69.2 | 58.4 | 67.6 | 66.2 | 53.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINOv2 + SDevaluation protocol=re-evaluated under identical standard evaluation settings from [20]2026.03 | 63.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 48.3 | 8.8 | — | — | — | — | — | — | — | — | — | — | — | — | |
| ACTRBackbone=iBOT, Resolution=256 x 2562026.01 | 62.1 | — | 42 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 44 | 4.9 | — | — | — | — | |
| CATS++supervision=Strong Supervised (S), protocol=per image PCK, fine-tuned backbone=true2023.05 | 59.9 | — | — | 60.6 | 46.9 | 82.5 | 41.6 | 56.8 | 64.9 | 50.4 | 72.8 | 29.2 | 75.8 | 65.4 | 62.5 | 50.9 | 56.1 | 54.8 | 48.2 | 80.9 | 74.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SD+DINOv2Supervision Type=Weakly Supervised without human annotations2026.05 | 59.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 44.7 | 7.9 | — | — | — | — | — | — | — | — | — | — | — | — | |
| SD+DINOv2Supervision level=Weakly supervised without human annotations2026.05 | 59.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 44.7 | 7.9 | — | — | — | — | — | — | — | — | — | — | — | — | |
| DistillDIFTevaluation protocol=re-evaluated under identical standard evaluation settings from [20]2026.03 | 59.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 42.7 | 5.7 | — | — | — | — | — | — | — | — | — | — | — | — | |
| SimpleMatchBackbone=ResNet101, Resolution=256 x 2562026.01 | 59.6 | — | 43.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 98 | 2.7 | — | — | — | — | |
| KBCNetBackbone=ResNet101, Resolution=256 x 2562026.01 | 59.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TransforMatcherBackbone=iBOT, Resolution=256 x 2562026.01 | 57.9 | — | 33.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Stable Diffusionsupervision=Unsupervised nearest neighboring (UN), protocol=per point PCK2023.05 | 57.2 | — | — | 63.1 | 55.6 | 80.2 | 33.8 | 44.9 | 49.3 | 47.8 | 74.4 | 38.4 | 70.8 | 53.7 | 61.1 | 54.4 | 55 | 54.8 | 53.5 | 65 | 53.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINOv2-ViT-B/14supervision=Unsupervised nearest neighboring (UN), protocol=per point PCK2023.05 | 55.6 | — | — | 72.7 | 62 | 85.2 | 41.3 | 40.4 | 52.3 | 51.5 | 71.1 | 36.2 | 67.1 | 64.6 | 67.6 | 61 | 68.2 | 30.7 | 62 | 54.3 | 24.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SCorrSANsupervision=Strong Supervised (S), protocol=per image PCK, fine-tuned backbone=true2023.05 | 55.3 | — | — | 57.1 | 40.3 | 78.3 | 38.1 | 51.8 | 57.8 | 47.1 | 67.9 | 25.2 | 71.3 | 63.9 | 49.3 | 45.3 | 49.8 | 48.8 | 40.3 | 77.7 | 69.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CATsBackbone=iBOT, Resolution=256 x 2562026.01 | 55.2 | — | 30.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINOv2+NNSupervision Type=Unsupervised2026.05 | 53.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 38.4 | 6.3 | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINOv2+NNSupervision level=Unsupervised2026.05 | 53.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 38.4 | 6.3 | — | — | — | — | — | — | — | — | — | — | — | — | |
| TransforMatcherBackbone=ResNet101, Resolution=240 x 2402026.01 | 53.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 133 | 2.7 | — | — | — | — | |
| DIFTSupervision Type=Unsupervised2026.05 | 52.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 39.7 | 7.2 | — | — | — | — | — | — | — | — | — | — | — | — | |
| DIFTSupervision level=Unsupervised2026.05 | 52.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 39.7 | 7.2 | — | — | — | — | — | — | — | — | — | — | — | — | |
| SemiMatchBackbone=ResNet-101, Supervision=Semi-sup., Learning signal=keypoints, Fine-tuning=true2022.03 | 50.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PMNCsupervision=Strong Supervised (S), protocol=per image PCK, fine-tuned backbone=true2023.05 | 50.4 | — | — | 54.1 | 35.9 | 74.9 | 36.5 | 42.1 | 48.8 | 40 | 72.6 | 21.1 | 67.6 | 58.1 | 50.5 | 40.1 | 54.1 | 43.3 | 35.7 | 74.5 | 59.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MMNet-FCNBackbone=ResNet101, Resolution=224 x 3202026.01 | 50.4 | — | 33.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 15 | 6.9 | — | — | — | — | |
| CATSBackbone=ResNet-101, Supervision=Sup., Learning signal=keypoints, Fine-tuning=true2022.03 | 49.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CATSsupervision=Strong Supervised (S), protocol=per image PCK, fine-tuned backbone=true2023.05 | 49.9 | — | — | 52 | 34.7 | 72.2 | 34.3 | 49.9 | 57.5 | 43.6 | 66.5 | 24.4 | 63.2 | 56.5 | 52 | 42.6 | 41.7 | 43 | 33.6 | 72.6 | 58 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CATsBackbone=ResNet101, Resolution=256 x 2562026.01 | 49.9 | — | 27.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 92 | 3 | — | — | — | — | |
| CHMNetBackbone=ResNet-101, Supervision=Sup., Learning signal=keypoints2022.03 | 46.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CHMBackbone=ResNet101, Resolution=240 x 2402026.01 | 46.3 | — | 22.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 41 | 2.6 | — | — | — | — | |
| SemiMatch+Backbone=ResNet-101, Supervision=Semi-sup., Learning signal=keypoints, Fine-tuning=false2022.03 | 43 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CATS+Backbone=ResNet-101, Supervision=Sup., Learning signal=keypoints, Fine-tuning=false2022.03 | 42.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MMNetBackbone=ResNet-101, Supervision=Sup., Learning signal=keypoints2022.03 | 40.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DHPFBackbone=ResNet-101, Supervision=Weak-sup., Learning signal=images2022.03 | 37.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DHPFBackbone=ResNet-101, Supervision=Sup., Learning signal=keypoints2022.03 | 37.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DHPFBackbone=ResNet101, Resolution=240 x 2402026.01 | 37.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 44 | 3.5 | — | — | — | — | |
| PWarpC-NC-Net*res101Resolution=ori, Supervision Level=Strong supervision2022.03 | 37.1 | — | 21.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ASICsupervision=Unsupervised task-specific (UT), protocol=per point PCK2023.05 | 36.9 | — | — | 57.9 | 25.2 | 68.1 | 24.7 | 35.4 | 28.4 | 30.9 | 54.8 | 21.6 | 45 | 47.2 | 39.9 | 26.2 | 48.8 | 14.5 | 24.5 | 49 | 24.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SCOTsupervision=Strong Supervised (S), protocol=per image PCK2023.05 | 35.6 | — | — | 34.9 | 20.7 | 63.8 | 21.1 | 43.5 | 27.3 | 21.3 | 63.1 | 20 | 42.9 | 42.5 | 31.1 | 29.8 | 35 | 27.7 | 24.4 | 48.4 | 40.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PWarpC-NC-Netres101Resolution=ori, Supervision Level=Weakly supervised2022.03 | 35.3 | — | 18.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GANgealingsupervision=GAN supervised (G), protocol=per point PCK2023.05 | 35.2 | — | — | 37.8 | — | — | — | — | — | — | 67 | — | — | 23.1 | — | — | — | — | — | 53.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| WarpC-NC-Netres101Resolution=ori, Supervision Level=Weakly supervised2022.03 | 35 | — | 18 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PWarpC-SF-Netres101Resolution=ori, Supervision Level=Weakly supervised2022.03 | 33.5 | — | 17.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GSFres101Resolution=-, Supervision Level=Weakly supervised2022.03 | 33.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINOv1-ViT-S/8supervision=Unsupervised nearest neighboring (UN), protocol=per point PCK2023.05 | 33.3 | — | — | 57.2 | 24.1 | 67.4 | 24.5 | 26.8 | 29 | 27.1 | 52.1 | 15.7 | 42.4 | 43.3 | 30.1 | 23.2 | 40.7 | 16.6 | 24.1 | 31 | 24.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PWarpC-SF-Net*res101Resolution=ori, Supervision Level=Strong supervision2022.03 | 32.5 | — | 17.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NC-Net*res101Resolution=ori, Supervision Level=Strong supervision2022.03 | 32.4 | — | 17.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINO+LiFTResolution=4482024.03 | 31.38 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 18.9 | 2.52 | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINO+MLSsupervision=Unsupervised task-specific (UT), protocol=per point PCK2023.05 | 31.1 | — | — | 49.7 | 20.9 | 63.9 | 19.1 | 32.5 | 27.6 | 22.4 | 48.9 | 14 | 36.9 | 39 | 30.1 | 21.7 | 41.1 | 17.1 | 18.1 | 35.9 | 21.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TransforMatcherSupervision=Strong, Backbone=ResNet-101, time (ms)=54, memory (GB)=1.6, FLOPs (G)=33.52022.05 | 30.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINO+BLResolution=4482024.03 | 30.48 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 16.75 | 1.64 | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINO+RCResolution=4482024.03 | 30.21 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 16.76 | 1.7 | — | — | — | — | — | — | — | — | — | — | — | — | |
| CHMNetSupervision=Strong, Backbone=ResNet-101, time (ms)=54, memory (GB)=1.6, FLOPs (G)=19.62022.05 | 30.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TransforMatcherSupervision=Strong, Backbone=ResNet-101, Data Augmentation=true, time (ms)=54, memory (GB)=1.6, FLOPs (G)=33.52022.05 | 30.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINO+JBUResolution=4482024.03 | 29.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 16.01 | 1.42 | — | — | — | — | — | — | — | — | — | — | — | — | |
| PMNCres101Resolution=-, Supervision Level=Strong supervision2022.03 | 28.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NC-Netres101Resolution=ori †, Supervision Level=Weakly supervised2022.03 | 28.8 | — | 13.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ANC-NetBackbone=ResNet-101-FCN, Supervision=Sup., Learning signal=keypoints2022.03 | 28.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ANC-NetSupervision=Strong, Backbone=ResNet-101, time (ms)=216, memory (GB)=0.9, FLOPs (G)=44.92022.05 | 28.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ANC-Netres101Resolution=-, Supervision Level=Strong supervision2022.03 | 28.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINO+LiFTResolution=2242024.03 | 28.68 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 14.72 | 0.91 | — | — | — | — | — | — | — | — | — | — | — | — | |
| PWarpC-DHPFres101Resolution=ori, Supervision Level=Strong supervision2022.03 | 28.6 | — | 16.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINOResolution=4482024.03 | 28.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 15.33 | 1.4 | — | — | — | — | — | — | — | — | — | — | — | — | |
| DHPFSupervision=Weak, Backbone=ResNet-101, time (ms)=58, memory (GB)=1.6, FLOPs (G)=22022.05 | 28.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DHPFres101Resolution=240, Supervision Level=Weakly supervised2022.03 | 28.5 | — | 14.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HPFBackbone=ResNet-101, Supervision=Sup., Learning signal=keypoints2022.03 | 28.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PWarpC-CATs-ft-featuresres101Resolution=ori, Supervision Level=Strong supervision2022.03 | 27.9 | — | 15.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SF-Net*res101Resolution=ori, Supervision Level=Strong supervision2022.03 | 27.9 | — | 13.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DHPFres101Resolution=ori †, Supervision Level=Weakly supervised2022.03 | 27.8 | — | 14.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DHPFres101Resolution=ori, Supervision Level=Strong supervision2022.03 | 27.5 | — | 15.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DHPFSupervision=Strong, Backbone=ResNet-101, time (ms)=58, memory (GB)=1.6, FLOPs (G)=22022.05 | 27.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DHPFres101Resolution=240, Supervision Level=Strong supervision2022.03 | 27.4 | — | 15.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| VGG+MLSsupervision=Unsupervised task-specific (UT), protocol=per point PCK2023.05 | 27.4 | — | — | 29.5 | 22.7 | 61.9 | 26.5 | 20.6 | 25.4 | 14.1 | 23.7 | 14.2 | 27.6 | 30 | 29.1 | 24.7 | 27.4 | 19.1 | 19.3 | 24.4 | 22.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SelfPatchResolution=4482024.03 | 27.34 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 14.44 | 1.12 | — | — | — | — | — | — | — | — | — | — | — | — | |
| CATsSupervision=Strong, Backbone=ResNet-101, Data Augmentation=true, time (ms)=45, memory (GB)=1.6, FLOPs (G)=28.42022.05 | 27.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — |