Image-Based Shape Retrieval on CompCars 50 (test)
10.9Top-1 Accuracy (instance)Point-BERT(L)
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Point-BERT(L)Framework=OpenShape, Pre-train Data=Ensembled, Zero-shot=true2026.03 | 10.9 | — | 44.2 | — | 19.5 | — | |
| Point-BERT(S)Framework=OpenShape, Pre-train Data=Shapenet, Zero-shot=true2026.03 | 10.8 | — | 47.7 | — | 20.1 | — | |
| SparseConvFramework=OpenShape, Pre-train Data=Ensembled, Zero-shot=true2026.03 | 7.7 | — | 29 | — | 12.7 | — | |
| SparseConvFramework=OpenShape, Pre-train Data=Shapenet, Zero-shot=true2026.03 | 3.8 | — | 22.6 | — | 8.1 | — | |
| Point-BERTFramework=ULIP, Pre-train Data=Ensembled, Zero-shot=true2026.03 | 1.1 | — | 10 | — | 2.9 | — | |
| Point-BERTFramework=ULIP, Pre-train Data=Shapenet, Zero-shot=true2026.03 | 0.7 | — | 10.1 | — | 2.6 | — | |
| Point-BERTFramework=ULIP2, Pre-train Data=Ensembled, Zero-shot=true2026.03 | 0.7 | — | 12.8 | — | 3.2 | — | |
| PointNeXtFramework=ULIP2, Pre-train Data=Ensembled, Zero-shot=true2026.03 | 0.5 | — | 11.3 | — | 3.1 | — |