3D Object Classification on ModelNet40 8k P v1
94.7Overall AccuracyReCon
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ReCon#P (M)=43.6, #F (G)=5.3, Training Paradigm=with Pretrained Cross-Modal Teacher Representation Learning2023.07 | 94.7 | — | — | |
| ReCon2025.12 | 94.7 | — | — | |
| VPP w/ vot.#P (M)=22.1, #F (G)=4.8, Training Paradigm=with Self-Supervised Representation Learning, Voting Strategy=Yes2023.07 | 94.3 | — | — | |
| Point-PQAE2025.12 | 94.3 | — | — | |
| CSCon2025.12 | 94.3 | — | — | |
| PointGPT2025.12 | 94.2 | — | — | |
| Point-MAE#P (M)=22.1, #F (G)=4.8, Training Paradigm=with Self-Supervised Representation Learning2023.07 | 94 | — | — | |
| VPP w/o vot.#P (M)=22.1, #F (G)=4.8, Training Paradigm=with Self-Supervised Representation Learning, Voting Strategy=No2023.07 | 94 | — | — | |
| ACT#P (M)=22.1, #F (G)=4.8, Training Paradigm=with Pretrained Cross-Modal Teacher Representation Learning2023.07 | 94 | — | — | |
| Point-MAE2025.12 | 94 | — | — | |
| ACT2025.12 | 94 | — | — | |
| Point-BERT#P (M)=22.1, #F (G)=4.8, Training Paradigm=with Self-Supervised Representation Learning2023.07 | 93.8 | — | — | |
| Point-BERT2025.12 | 93.8 | — | — | |
| PointNet++#P (M)=1.5, #F (G)=1.7, Training Paradigm=Supervised Learning Only2023.07 | 91.9 | — | — | |
| PointNet++2025.12 | 91.9 | — | — | |
| Transformer#P (M)=22.1, #F (G)=4.8, Training Paradigm=with Self-Supervised Representation Learning2023.07 | 91.8 | — | — | |
| PointNet#P (M)=3.5, #F (G)=0.5, Training Paradigm=Supervised Learning Only2023.07 | 90.8 | — | — | |
| PointNet2025.12 | 90.8 | — | — | |
| ACTReference=ICLR 23, Params. (M)=22.1, Training Protocol=Self-Supervised Representation Learning (FULL)2024.08 | — | 94 | — | |
| DAPTReference=CVPR 24, Params. (M)=1.1, Baseline=Point-MAE, Training Protocol=Parameter-Efficient Supervised Fine-tuning2024.08 | — | 93.27 | 93.05 | |
| DAPTReference=CVPR 24, Params. (M)=1.1, Baseline=RECON, Training Protocol=Parameter-Efficient Supervised Fine-tuning2024.08 | — | 93.07 | 92.79 | |
| IDPTReference=ICCV 23, Params. (M)=1.7, Baseline=Point-MAE, Training Protocol=Parameter-Efficient Supervised Fine-tuning2024.08 | — | 93.88 | 93.67 | |
| IDPTReference=ICCV 23, Params. (M)=1.7, Baseline=RECON, Training Protocol=Parameter-Efficient Supervised Fine-tuning2024.08 | — | 93.64 | 93.52 | |
| OcCoReference=ICCV 21, Params. (M)=22.1, Training Protocol=Self-Supervised Representation Learning (FULL)2024.08 | — | — | 92.1 | |
| Point-BERTReference=CVPR 22, Params. (M)=22.1, Training Protocol=Self-Supervised Representation Learning (FULL)2024.08 | — | 93.8 | — | |
| Point-MAEReference=ECCV 22, Params. (M)=22.1, Training Protocol=Self-Supervised Representation Learning (FULL)2024.08 | — | 94 | — | |
| Point-MAE (Baseline)Reference=ECCV 22, Params. (M)=22.1, Training Protocol=Parameter-Efficient Supervised Fine-tuning2024.08 | — | 94 | — | |
| PointNetReference=CVPR 17, Params. (M)=3.5, Training Protocol=Supervised Learning Only2024.08 | — | 90.8 | — | |
| PointNet++Reference=NeurIPS 17, Params. (M)=1.5, Training Protocol=Supervised Learning Only2024.08 | — | 91.9 | — | |
| PPTReference=ACMMM 25, Params. (M)=1.1, Baseline=Point-MAE, Training Protocol=Parameter-Efficient Supervised Fine-tuning2024.08 | — | 93.88 | 93.51 | |
| PPTReference=ACMMM 25, Params. (M)=1.1, Baseline=RECON, Training Protocol=Parameter-Efficient Supervised Fine-tuning2024.08 | — | 93.84 | 93.66 | |
| RECONReference=ICML 23, Params. (M)=44.3, Training Protocol=Self-Supervised Representation Learning (FULL)2024.08 | — | 94.3 | — | |
| RECON (Baseline)Reference=ICML 23, Params. (M)=22.1, Training Protocol=Parameter-Efficient Supervised Fine-tuning2024.08 | — | 93 | — | |
| VPPReference=NeurIPS 23, Params. (M)=22.1, Training Protocol=Self-Supervised Representation Learning (FULL)2024.08 | — | 94.3 | — |