3D Point Cloud Classification on ModelNet40
94.7AccuracyAsymDSD-B*
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| AsymDSD-B*Evaluation Protocol=Full Fine-tune, PP (Post-pretraining)=false, CM (Cross-modal training)=false, #P(M)=92.12025.06 | 94.7 | — | — | — | |
| ReCon++-B*Evaluation Protocol=Full Fine-tune, PP (Post-pretraining)=true, CM (Cross-modal training)=true, #P(M)=177.42025.06 | 94.6 | — | — | — | |
| PointGPT-LEvaluation Protocol=Full Fine-tune, PP (Post-pretraining)=false, CM (Cross-modal training)=false, #P(M)=310.02025.06 | 94.5 | — | — | — | |
| AsymDSD-S*Evaluation Protocol=Full Fine-tune, PP (Post-pretraining)=false, CM (Cross-modal training)=false, #P(M)=22.12025.06 | 94.4 | — | — | — | |
| PointGPT-B*Evaluation Protocol=Full Fine-tune, PP (Post-pretraining)=true, CM (Cross-modal training)=false, #P(M)=92.02025.06 | 94.4 | — | — | — | |
| PointGPT-BEvaluation Protocol=Full Fine-tune, PP (Post-pretraining)=false, CM (Cross-modal training)=false, #P(M)=92.02025.06 | 94.2 | — | — | — | |
| Point-MAE-B*Evaluation Protocol=Full Fine-tune, PP (Post-pretraining)=true, CM (Cross-modal training)=false, #P(M)=120.12025.06 | 94.2 | — | — | — | |
| PointMLP#P(M)=12.6, ST (Standard Transformer)=true, SM (Single-modal training)=true, Evaluation Protocol=Supervised2025.06 | 94.1 | — | — | — | |
| AsymDSD-S#P(M)=22.1, ST (Standard Transformer)=true, SM (Single-modal training)=true, Evaluation Protocol=Full Fine-tune2025.06 | 94.1 | — | — | — | |
| PVTParams=2.76M, FLOPs=1.93G2021.08 | 94.1 | — | — | 9.1 | |
| PointNeXt#P(M)=1.4, ST (Standard Transformer)=false, SM (Single-modal training)=true, Evaluation Protocol=Supervised2025.06 | 94 | — | — | — | |
| PointGPT-S#P(M)=22.1, ST (Standard Transformer)=true, SM (Single-modal training)=true, Evaluation Protocol=Full Fine-tune2025.06 | 94 | — | — | — | |
| AsymSD-MPM-S#P(M)=22.1, ST (Standard Transformer)=true, SM (Single-modal training)=true, Evaluation Protocol=Full Fine-tune2025.06 | 94 | — | — | — | |
| Point-M2AE#P(M)=15.3, ST (Standard Transformer)=false, SM (Single-modal training)=true, Evaluation Protocol=Full Fine-tune2025.06 | 94 | — | — | — | |
| Point-RAE#P(M)=29.2, ST (Standard Transformer)=false, SM (Single-modal training)=true, Evaluation Protocol=Full Fine-tune2025.06 | 94 | — | — | — | |
| Point-FEMAE#P(M)=27.4, ST (Standard Transformer)=false, SM (Single-modal training)=true, Evaluation Protocol=Full Fine-tune2025.06 | 94 | — | — | — | |
| AsymDSD-S*Evaluation Protocol=Linear, PP (Post-pretraining)=false, CM (Cross-modal training)=false, #P(M)=21.82025.06 | 93.98 | — | — | — | |
| MaskPointST (Standard Transformer)=false, SM (Single-modal training)=true, Evaluation Protocol=Full Fine-tune2025.06 | 93.8 | — | — | — | |
| AsymDSD-B*Evaluation Protocol=Linear, PP (Post-pretraining)=false, CM (Cross-modal training)=false, #P(M)=91.82025.06 | 93.76 | — | — | — | |
| PT2Params=9.14M, FLOPs=17.1G2021.08 | 93.7 | — | — | 65.4 | |
| AsymSD-CLS-S#P(M)=22.1, ST (Standard Transformer)=true, SM (Single-modal training)=true, Evaluation Protocol=Full Fine-tune2025.06 | 93.6 | — | — | — | |
| ReCon SM#P(M)=43.6, ST (Standard Transformer)=false, SM (Single-modal training)=true, Evaluation Protocol=Full Fine-tune2025.06 | 93.6 | — | — | — | |
| PointMamba#P(M)=12.3, ST (Standard Transformer)=false, SM (Single-modal training)=true, Evaluation Protocol=Full Fine-tune2025.06 | 93.6 | — | — | — | |
| AsymSD-MPM-S#P(M)=21.8, ST (Standard Transformer)=true, SM (Single-modal training)=true, Evaluation Protocol=Linear2025.06 | 93.55 | — | — | — | |
| Point-BERT#P(M)=22.1, ST (Standard Transformer)=true, SM (Single-modal training)=true, Evaluation Protocol=Full Fine-tune2025.06 | 93.2 | — | — | — | |
| Point-MAE#P(M)=22.1, ST (Standard Transformer)=true, SM (Single-modal training)=true, Evaluation Protocol=Full Fine-tune2025.06 | 93.2 | — | — | — | |
| PCTParams=2.88M, FLOPs=2.17G2021.08 | 93.2 | — | — | 24.6 | |
| PointASNLParams=3.98M, FLOPs=5.92G2021.08 | 93.1 | — | — | 39.8 | |
| Adapted ViT-S#P(M)=22.1, ST (Standard Transformer)=true, SM (Single-modal training)=true, Evaluation Protocol=Supervised2025.06 | 92.9 | — | — | — | |
| RSCNN-Multi# Points=1024, Closest Protocol=RSCNN no Vote2021.06 | 92.9 | — | — | — | |
| PointANSL# Points=1024, Closest Protocol=DGCNN CE*2021.06 | 92.9 | — | — | — | |
| DGCNNParams=1.81M, FLOPs=2.43G2021.08 | 92.9 | — | — | 57.2 | |
| DensePoint# Points=1024, Closest Protocol=RSCNN no Vote2021.06 | 92.8 | — | — | — | |
| PT1Params=21.1M, FLOPs=5.05G2021.08 | 92.8 | — | — | 32.5 | |
| AsymDSD-S#P(M)=21.8, ST (Standard Transformer)=true, SM (Single-modal training)=true, Evaluation Protocol=Linear2025.06 | 92.52 | — | — | — | |
| PointCNN# Points=1024, Closest Protocol=DGCNN CE*2021.06 | 92.5 | — | — | — | |
| ReCon#P(M)=43.3, ST (Standard Transformer)=false, SM (Single-modal training)=false, Evaluation Protocol=Linear2025.06 | 92.47 | — | — | — | |
| PCNN# Points=1024, Closest Protocol=DCNN CE Final*2021.06 | 92.3 | — | — | — | |
| PointNet++(MSG)Params=1.74M, FLOPs=4.09G2021.08 | 91.9 | — | — | 47.6 | |
| AsymSD-CLS-S#P(M)=21.8, ST (Standard Transformer)=true, SM (Single-modal training)=true, Evaluation Protocol=Linear2025.06 | 91.78 | — | — | — | |
| 3DmFV# Points=1024, Closest Protocol=DGCNN CE*2021.06 | 91.4 | — | — | — | |
| ACT#P(M)=21.8, ST (Standard Transformer)=true, SM (Single-modal training)=false, Evaluation Protocol=Linear2025.06 | 91.36 | — | — | — | |
| SO-Net# Points=2048, Closest Protocol=DGCNN CE*2021.06 | 90.9 | — | — | — | |
| PointNet++(SSG)Params=1.48M, FLOPs=1.68G2021.08 | 90.7 | — | — | 43.5 | |
| Recon MAE#P(M)=43.3, ST (Standard Transformer)=false, SM (Single-modal training)=true, Evaluation Protocol=Linear2025.06 | 90.22 | — | — | — | |
| MVCNN, f.t.Pre-train=ImageNet1K, Fine-tune=ModelNet40, Training #Views=80, Test #Views=802015.05 | 90.1 | — | — | — | |
| MVCNN, f.t. + metricPre-train=ImageNet1K, Fine-tune=ModelNet40, Training #Views=80, Test #Views=80, Low-rank Mahalanobis metric=true2015.05 | 90.1 | — | — | — | |
| DeepSets# Points=5000, Closest Protocol=PointNet++ no Vote*2021.06 | 90 | — | — | — | |
| MVCNN, f.t.Pre-train=ImageNet1K, Fine-tune=ModelNet40, Training #Views=12, Test #Views=122015.05 | 89.9 | — | — | — | |
| MVCNN, f.t. + metricPre-train=ImageNet1K, Fine-tune=ModelNet40, Training #Views=12, Test #Views=12, Low-rank Mahalanobis metric=true2015.05 | 89.5 | — | — | — | |
| PointNet#P(M)=3.5, ST (Standard Transformer)=false, SM (Single-modal training)=true, Evaluation Protocol=Supervised2025.06 | 89.2 | — | — | — | |
| PointNetParams=3.47M, FLOPs=0.45G2021.08 | 89.2 | — | — | 0 | |
| CNN, f.t.Pre-train=ImageNet1K, Fine-tune=ModelNet40, Training #Views=12, Test #Views=122015.05 | 88.6 | — | — | — | |
| MVCNNPre-train=ImageNet1K, Test #Views=122015.05 | 88.1 | — | — | — | |
| CNNPre-train=ImageNet1K, Test #Views=122015.05 | 87.5 | — | — | — | |
| CNN, f.t.Pre-train=ImageNet1K, Fine-tune=ModelNet40, Training #Views=12, Test #Views=12015.05 | 85.1 | — | — | — | |
| FVFine-tune=ModelNet40, Training #Views=12, Test #Views=122015.05 | 84.8 | — | — | — | |
| MVCNNPre-train=ImageNet1K, Training #Views=80, Test #Views=802015.05 | 84.3 | — | — | — | |
| CNNPre-train=ImageNet1K, Test #Views=12015.05 | 83 | — | — | — | |
| FVFine-tune=ModelNet40, Training #Views=12, Test #Views=12015.05 | 78.8 | — | — | — | |
| 3D ShapeNetsPre-train=ModelNet40, Fine-tune=ModelNet402015.05 | 77.3 | — | — | — | |
| LFD2015.05 | 75.5 | — | — | — | |
| SPH2015.05 | 68.2 | — | — | — | |
| DGCNNTraining Dataset=ScanObjectNN2021.06 | 54.7 | — | — | — | |
| 3DmFVTraining Dataset=ScanObjectNN2021.06 | 51.5 | — | — | — | |
| PointNetTraining Dataset=ScanObjectNN2021.06 | 50.9 | — | — | — | |
| PointCNNTraining Dataset=ScanObjectNN2021.06 | 49.2 | — | — | — | |
| PointNet++Training Dataset=ScanObjectNN2021.06 | 47.4 | — | — | — | |
| SpiderCNNTraining Dataset=ScanObjectNN2021.06 | 46.6 | — | — | — | |
| PointNet++# Points=5000, Closest Protocol=PointNet++ no Vote*2021.06 | — | 89.8 | — | — | |
| PointNet++# Points=2048, Closest Protocol=DGCNN CE*2021.06 | — | 92.6 | — | — | |
| PointNet++# Points=1024, Closest Protocol=DGCNN CE*2021.06 | — | 92.6 | — | — | |
| PointNet++# Points=1024, Closest Protocol=DCNN CE Final*2021.06 | — | 92.1 | — | — | |
| PointNet++# Points=1024, Closest Protocol=DGCNN CE*2021.06 | — | 92.6 | — | — | |
| PointNet++# Points=1024, Closest Protocol=RSCNN no Vote2021.06 | — | 92.7 | — | — | |
| PointNet++# Points=1024, Closest Protocol=RSCNN no Vote2021.06 | — | 92.7 | — | — | |
| PointNet++# Points=1024, Closest Protocol=DGCNN CE*2021.06 | — | 92.6 | — | — | |
| SimpleView# Points=5000, Closest Protocol=PointNet++ no Vote*2021.06 | — | — | 90.7 | — | |
| SimpleView# Points=2048, Closest Protocol=DGCNN CE*2021.06 | — | — | 93.1 | — | |
| SimpleView# Points=1024, Closest Protocol=DGCNN CE*2021.06 | — | — | 93.1 | — | |
| SimpleView# Points=1024, Closest Protocol=DCNN CE Final*2021.06 | — | — | 92.5 | — | |
| SimpleView# Points=1024, Closest Protocol=DGCNN CE*2021.06 | — | — | 93.1 | — | |
| SimpleView# Points=1024, Closest Protocol=RSCNN no Vote2021.06 | — | — | 92.9 | — | |
| SimpleView# Points=1024, Closest Protocol=RSCNN no Vote2021.06 | — | — | 92.9 | — | |
| SimpleView# Points=1024, Closest Protocol=DGCNN CE*2021.06 | — | — | 93.1 | — |