Object Identification on Object Folder Real
92.7Top-1 AccuracyViTaPEs
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ViTaPEsModel Backbone=ViT, SSL=false, # params=12.7M2025.05 | 92.7 | — | |
| Vanilla CNNModel Backbone=ResNet, SSL=false, # params=22.3M2025.05 | 89.8 | — | |
| ViTaPEsModel Backbone=ViT, SSL=true, # params=30.6M2025.05 | 85.2 | — | |
| RoPEModel Backbone=ViT, SSL=false, # params=12.0M2025.05 | 84.7 | — | |
| VTTModel Backbone=ViT, SSL=false, # params=12.0M2025.05 | 83.6 | — | |
| MViTacModel Backbone=ResNet, SSL=true, # params=22.4M2025.05 | 82.3 | — | |
| TAGModel Backbone=ResNet, SSL=true, # params=22.3M2025.05 | 81.2 | — | |
| RoPEModel Backbone=ViT, SSL=true, # params=29.7M2025.05 | 77.5 | — | |
| VTTModel Backbone=ViT, SSL=true, # params=29.7M2025.05 | 76.8 | — | |
| SSVTPModel Backbone=ResNet, SSL=true, # params=22.4M2025.05 | 53.8 | — | |
| Patch VisTac Late Fusionmodality=Vision-Tactile, input_type=Patch, fusion=Late2026.06 | 35.43 | 78.2 | |
| Patch VisTac Early Fusionmodality=Vision-Tactile, input_type=Patch, fusion=Early2026.06 | 33.36 | 73.87 | |
| Patch Vis + CLmodality=Vision, input_type=Patch, alignment=CL2026.06 | 33.25 | — | |
| Patch Vismodality=Vision, input_type=Patch2026.06 | 32.83 | — | |
| Patch VisTac Early Fusion+CLmodality=Vision-Tactile, input_type=Patch, alignment=CL, fusion=Early2026.06 | 32.82 | 70.28 | |
| Patch VisTac Late Fusionmodality=Vision-Tactile, input_type=Patch, fusion=Late2026.06 | 31.54 | — | |
| Patch Vis + CLmodality=Vision, input_type=Patch, alignment=CL2026.06 | 31.27 | 71.21 | |
| Patch Tac + CLmodality=Tactile, input_type=Patch, alignment=CL2026.06 | 31.21 | — | |
| Patch Vis + DSCMRmodality=Vision, input_type=Patch, alignment=DSCMR2026.06 | 30.77 | — | |
| Patch VisTac Late Fusion+CLmodality=Vision-Tactile, input_type=Patch, alignment=CL, fusion=Late2026.06 | 30.43 | 76.2 | |
| Patch Tac + DSCMRmodality=Tactile, input_type=Patch, alignment=DSCMR2026.06 | 30.39 | — | |
| Patch VisTac Early Fusion+DARmodality=Vision-Tactile, input_type=Patch, alignment=DAR, fusion=Early2026.06 | 30.37 | — | |
| Vis + GAmodality=Vision, alignment=GA2026.06 | 30.14 | — | |
| Patch VisTac Late Fusion+CLmodality=Vision-Tactile, input_type=Patch, alignment=CL, fusion=Late2026.06 | 29.32 | — | |
| Vismodality=Vision2026.06 | 29.21 | — | |
| Patch VisTac Early Fusionmodality=Vision-Tactile, input_type=Patch, fusion=Early2026.06 | 29.03 | — | |
| Patch Vis + DSCMRmodality=Vision, input_type=Patch, alignment=DSCMR2026.06 | 28.79 | 60.68 | |
| Tacmodality=Tactile2026.06 | 28.47 | — | |
| Patch Tacmodality=Tactile, input_type=Patch2026.06 | 28.44 | — | |
| Patch Vismodality=Vision, input_type=Patch2026.06 | 27.86 | 65.63 | |
| Patch Vis + DARmodality=Vision, input_type=Patch, alignment=DAR2026.06 | 27.85 | — | |
| Tac + GAmodality=Tactile, alignment=GA2026.06 | 27.41 | — | |
| Patch VisTac Early Fusion+DSCMRmodality=Vision-Tactile, input_type=Patch, alignment=DSCMR, fusion=Early2026.06 | 27.12 | — | |
| Patch VisTac Late Fusion+DSCMRmodality=Vision-Tactile, input_type=Patch, alignment=DSCMR, fusion=Late2026.06 | 26.13 | — | |
| Patch Tac + DARmodality=Tactile, input_type=Patch, alignment=DAR2026.06 | 25.28 | — | |
| Patch VisTac Early Fusion+CLmodality=Vision-Tactile, input_type=Patch, alignment=CL, fusion=Early2026.06 | 24.61 | — | |
| Patch VisTac Late Fusion+DARmodality=Vision-Tactile, input_type=Patch, alignment=DAR, fusion=Late2026.06 | 24.18 | — | |
| Patch VisTac Early Fusion+DARmodality=Vision-Tactile, input_type=Patch, alignment=DAR, fusion=Early2026.06 | 23.84 | 67.49 | |
| Patch Tac + DSCMRmodality=Tactile, input_type=Patch, alignment=DSCMR2026.06 | 23.53 | 58.82 | |
| Patch VisTac Late Fusion+DARmodality=Vision-Tactile, input_type=Patch, alignment=DAR, fusion=Late2026.06 | 23.53 | 46.75 | |
| Vis + GAmodality=Vision, alignment=GA2026.06 | 22.29 | 68.42 | |
| Patch VisTac Early Fusion+DSCMRmodality=Vision-Tactile, input_type=Patch, alignment=DSCMR, fusion=Early2026.06 | 21.05 | 63.16 | |
| Vismodality=Vision2026.06 | 19.81 | 68.42 | |
| Patch Vis + DARmodality=Vision, input_type=Patch, alignment=DAR2026.06 | 18.27 | 68.73 | |
| Patch VisTac Late Fusion+DSCMRmodality=Vision-Tactile, input_type=Patch, alignment=DSCMR, fusion=Late2026.06 | 17.65 | 57.89 | |
| Patch Tac + CLmodality=Tactile, input_type=Patch, alignment=CL2026.06 | 16.64 | 60.15 | |
| Tacmodality=Tactile2026.06 | 14.24 | 49.54 | |
| Patch Tacmodality=Tactile, input_type=Patch2026.06 | 13.93 | 54.18 | |
| Patch Tac + DARmodality=Tactile, input_type=Patch, alignment=DAR2026.06 | 13.93 | 53.56 | |
| Tac + GAmodality=Tactile, alignment=GA2026.06 | 10.22 | 53.25 | |
| Chance2026.06 | 6.09 | — | |
| Chance2026.06 | 1.88 | 13.78 |