3D Object Recognition on ScanObjectNN
0.654Top-1 AccuracyRECON++-L
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| RECON++-LPre-training Data=Ensembled, Scale=Large2024.02 | 0.654 | 0.841 | 0.897 | |
| Uni3D-BPre-training Data=Ensembled, Teacher=EVA-CLIP-E2024.02 | 0.638 | 0.827 | 0.902 | |
| RECON++-BPre-training Data=Ensembled, Scale=Base2024.02 | 0.636 | 0.802 | 0.906 | |
| MixCon3DPre-training Data=Ensembled2024.02 | 0.586 | 0.803 | 0.892 | |
| Uni3D-LPre-training Data=Ensembled, Teacher=EVA-CLIP-E2024.02 | 0.582 | 0.818 | 0.894 | |
| TAMMPre-training Data=Ensembled2024.02 | 0.557 | 0.807 | 0.889 | |
| TAMMPre-training Data=ShapeNet2024.02 | 0.548 | 0.745 | 0.833 | |
| MixCon3DPre-training Data=ShapeNet2024.02 | 0.526 | 0.699 | 0.787 | |
| OpenShapePre-training Data=Ensembled2024.02 | 0.522 | 0.797 | 0.887 | |
| ULIP-2Pre-training Data=Ensembled2024.02 | 0.516 | 0.725 | 0.823 | |
| ULIPPre-training Data=ShapeNet2024.02 | 0.515 | 0.711 | 0.802 | |
| OpenShapePre-training Data=ShapeNet2024.02 | 0.472 | 0.724 | 0.847 | |
| RECONPre-training Data=ShapeNet2024.02 | 0.423 | 0.625 | 0.756 | |
| PointCLIPv2Protocol=2D Inference without 3D Training2024.02 | 0.422 | 0.633 | 0.745 | |
| CLIP2PointPre-training Data=ShapeNet2024.02 | 0.255 | 0.446 | 0.594 | |
| PointCLIPProtocol=2D Inference without 3D Training2024.02 | 0.105 | 0.208 | 0.306 |