# Object Classification on ScanObjectNN OBJ\_BG

Source: [Object Classification on ScanObjectNN OBJ\_BG](<https://www.sota2.com/research/sota/object-classification-on-scanobjectnn-obj_bg>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-19T10:55:24.359Z

Evaluation of Object Classification performance on the ScanObjectNN OBJ\_BG dataset, typically measured by Overall Accuracy.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Object Classification |
| Dataset (including any stated split) | ScanObjectNN OBJ\_BG |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 252 of 252 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Accuracy](<https://www.sota2.com/research/sota/object-classification-on-scanobjectnn-obj_bg.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Top-1 Acc](<https://www.sota2.com/research/sota/object-classification-on-scanobjectnn-obj_bg.md?metric=Top-1%20Acc>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| PointGST | 98.97 | Not recorded | 68fa7a3a1e209131cea81f2a |
| ReCon++-B\* | 98.62 | Not recorded | 68fa79b41e209131cea7efc4 |
| IDPT | 98.11 | Not recorded | 68fa7a3a1e209131cea81f2a |
| DAPT | 98.11 | Not recorded | 68fa7a3a1e209131cea81f2a |
| Fully fine-tune | 97.2 | Not recorded | 68fa7a3a1e209131cea81f2a |
| AsymDSD-S\* | 97.07 | Not recorded | 68fa79b41e209131cea7efc4 |
| AsymDSD-B\* | 96.73 | Not recorded | 68fa79b41e209131cea7efc4 |
| Point-PEFT | 96.39 | Not recorded | 68fa7a3a1e209131cea81f2a |
| AsymDSD-B\* | 96.16 | Not recorded | 68fa79b41e209131cea7efc4 |
| PointGPT-B\* | 95.8 | Not recorded | 68fa79b41e209131cea7efc4 |
| PointGPT-L | 95.7 | Not recorded | 68fa79b41e209131cea7efc4 |
| Point-RAE | 95.53 | Not recorded | 68fa79b41e209131cea7efc4 |
| Point-SRA | 95.53 | Not recorded | 695c970f53fdd5c361743838 |
| AsymDSD-S\* | 95.22 | Not recorded | 68fa79b41e209131cea7efc4 |
| Point-FEMAE | 95.18 | Not recorded | 68fa79b41e209131cea7efc4 |
| ReCon | 95.18 | Not recorded | 68fa70e91e209131cea4de6b |
| Mamba3D | 95.18 | Not recorded | 68fa74ae1e209131cea632e2 |
| Point-FEMAE | 95.18 | Not recorded | 695c970f53fdd5c361743838 |
| ReCon | 95.18 | Not recorded | 695c970f53fdd5c361743838 |
| Utonia | 95 | Not recorded | 69a7bca02d603e6de3e3d71e |
| PointGST | 94.49 | Not recorded | 68fa7a3a1e209131cea81f2a |
| Mamba3D | 94.49 | Not recorded | 68fa74ae1e209131cea632e2 |
| 3D-JEPA | 94.49 | Not recorded | 68fa74e61e209131cea64868 |
| AsymDSD-S | 94.32 | Not recorded | 68fa79b41e209131cea7efc4 |
| PointMamba | 94.32 | Not recorded | 68fa79b41e209131cea7efc4 |
| Fully fine-tune | 94.32 | Not recorded | 68fa7a3a1e209131cea81f2a |
| DAPT | 94.32 | Not recorded | 68fa7a3a1e209131cea81f2a |
| PointMamba | 94.32 | Not recorded | 695c970f53fdd5c361743838 |
| Point-MAE-B\* | 94.2 | Not recorded | 68fa79b41e209131cea7efc4 |
| ReCon SM | 94.15 | Not recorded | 68fa79b41e209131cea7efc4 |
| I2P-MAE | 94.15 | Not recorded | 68fa70e91e209131cea4de6b |
| RECON | 94.15 | Not recorded | 68fa78dd1e209131cea7ae14 |
| I2P-MAE | 94.15 | Not recorded | 695c970f53fdd5c361743838 |
| 3D-JEPA | 93.8 | Not recorded | 68fa74e61e209131cea64868 |
| PointNTP | 93.8 | Not recorded | 6a0beee6dbc836b87d82c7f0 |
| Point-MAE† w/ IDPT | 93.63 | Not recorded | 68fa70231e209131cea49508 |
| PointGPT-B | 93.6 | Not recorded | 68fa79b41e209131cea7efc4 |
| PointGST | 93.46 | Not recorded | 68fa7a3a1e209131cea81f2a |
| Standard Transformer | 93.29 | Not recorded | 68fa79651e209131cea7d694 |
| Fully fine-tune | 93.29 | Not recorded | 68fa7a3a1e209131cea81f2a |
| IDPT | 93.29 | Not recorded | 68fa7a3a1e209131cea81f2a |
| ACT | 93.29 | Not recorded | 68fa70231e209131cea49508 |
| ACT | 93.29 | Not recorded | 68fa70e91e209131cea4de6b |
| ACT | 93.29 | Not recorded | 68fa74e61e209131cea64868 |
| ACT | 93.29 | Not recorded | 68fa78dd1e209131cea7ae14 |
| PointDif | 93.29 | Not recorded | 695c970f53fdd5c361743838 |
| ACT | 93.29 | Not recorded | 695c970f53fdd5c361743838 |
| PointDif | 93.29 | Not recorded | 6a0beee6dbc836b87d82c7f0 |
| ACT | 93.29 | Not recorded | 6a0beee6dbc836b87d82c7f0 |
| Point-JEPA | 93.2 | Not recorded | 695c970f53fdd5c361743838 |
| IDPT | 93.12 | Not recorded | 68fa7a3a1e209131cea81f2a |
| ACT w/ IDPT | 93.12 | Not recorded | 68fa70231e209131cea49508 |
| Mamba3D | 93.12 | Not recorded | 68fa74ae1e209131cea632e2 |
| VPP w/ vot. | 93.11 | Not recorded | 68fa70e91e209131cea4de6b |
| Point-MAE† | 92.94 | Not recorded | 68fa70231e209131cea49508 |
| Mamba3D | 92.94 | Not recorded | 68fa74ae1e209131cea632e2 |
| AsymSD-CLS-S | 92.77 | Not recorded | 68fa79b41e209131cea7efc4 |
| AsymSD-MPM-S | 92.77 | Not recorded | 68fa79b41e209131cea7efc4 |
| VPP w/o vot. | 92.77 | Not recorded | 68fa70e91e209131cea4de6b |
| Standard Transformer | 92.6 | Not recorded | 68fa79651e209131cea7d694 |
| DAPT | 92.6 | Not recorded | 68fa7a3a1e209131cea81f2a |
| MVTN | 92.6 | Not recorded | 68fa74ae1e209131cea632e2 |
| MVTN | 92.6 | Not recorded | 6a0beee6dbc836b87d82c7f0 |
| UniPre3D (Std. Transformer) | 92.6 | Not recorded | 6a0beee6dbc836b87d82c7f0 |
| IAE (M2AE) | 92.5 | Not recorded | 6937ad1b6cfd488761a4e8a5 |
| IAE (M2AE) | 92.5 | Not recorded | 6a0beee6dbc836b87d82c7f0 |
| Concerto | 92.3 | Not recorded | 69a7bca02d603e6de3e3d71e |
| IAE (M2AE) | 92.3 | Not recorded | 6937ad1b6cfd488761a4e8a5 |
| Mamba3D | 92.25 | Not recorded | 68fa74ae1e209131cea632e2 |
| Sonata | 92.2 | Not recorded | 69a7bca02d603e6de3e3d71e |
| Point-PEFT | 91.91 | Not recorded | 68fa7a3a1e209131cea81f2a |
| RI-MAE | 91.9 | Not recorded | 695c970f53fdd5c361743838 |
| PointGST | 91.74 | Not recorded | 68fa7a3a1e209131cea81f2a |
| PointGPT-S | 91.6 | Not recorded | 68fa79b41e209131cea7efc4 |
| PointGPT-S | 91.6 | Not recorded | 6a0beee6dbc836b87d82c7f0 |
| PointGST | 91.39 | Not recorded | 68fa7a3a1e209131cea81f2a |
| Point-M2AE | 91.22 | Not recorded | 68fa79b41e209131cea7efc4 |
| IDPT | 91.22 | Not recorded | 68fa7a3a1e209131cea81f2a |
| Point-M2AE | 91.22 | Not recorded | 68fa70231e209131cea49508 |
| Point-MAE w/ IDPT | 91.22 | Not recorded | 68fa70231e209131cea49508 |

### PointGST

Accuracy: 98.97. Result ID: 698c576e37894ef7d9cda6d2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Implementation](<https://github.com/jerryfeng2003/PointGST>)

Recorded evaluation configuration: {"Pre-trained model":"PointGPT-L","Params. (M)":"2.4","FLOPS (G)":"67.95"}. This record does not assert independent reproduction.

### ReCon++-B\*

Accuracy: 98.62. Result ID: 698c537737894ef7d9cd8c8c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa740c1e209131cea5f73a.md>)

[Implementation](<https://qizekun.github.io/ShapeLLM>)

Recorded evaluation configuration: {"Evaluation Protocol":"Full Fine-tune","PP (Post-pretraining)":true,"CM (Cross-modal training)":true,"#P(M)":"177.4"}. This record does not assert independent reproduction.

### IDPT

Accuracy: 98.11. Result ID: 698c576e37894ef7d9cda6cf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70231e209131cea49508.md>)

[Implementation](<https://github.com/zyh16143998882/ICCV23-IDPT>)

Recorded evaluation configuration: {"Pre-trained model":"PointGPT-L","Params. (M)":"10.0","FLOPS (G)":"75.19"}. This record does not assert independent reproduction.

### DAPT

Accuracy: 98.11. Result ID: 698c576e37894ef7d9cda6d1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72d31e209131cea58c2e.md>)

Recorded evaluation configuration: {"Pre-trained model":"PointGPT-L","Params. (M)":"4.2","FLOPS (G)":"71.64"}. This record does not assert independent reproduction.

### Fully fine-tune

Accuracy: 97.2. Result ID: 698c576e37894ef7d9cda6ce.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

Recorded evaluation configuration: {"Pre-trained model":"PointGPT-L","Params. (M)":"360.5","FLOPS (G)":"67.71"}. This record does not assert independent reproduction.

### AsymDSD-S\*

Accuracy: 97.07. Result ID: 698c537737894ef7d9cd8c88.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Implementation](<https://github.com/RFLeijenaar/AsymDSD>)

Recorded evaluation configuration: {"Evaluation Protocol":"Full Fine-tune","PP (Post-pretraining)":false,"CM (Cross-modal training)":false,"#P(M)":"22.1"}. This record does not assert independent reproduction.

### AsymDSD-B\*

Accuracy: 96.73. Result ID: 698c537737894ef7d9cd8c89.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Implementation](<https://github.com/RFLeijenaar/AsymDSD>)

Recorded evaluation configuration: {"Evaluation Protocol":"Full Fine-tune","PP (Post-pretraining)":false,"CM (Cross-modal training)":false,"#P(M)":"92.1"}. This record does not assert independent reproduction.

### Point-PEFT

Accuracy: 96.39. Result ID: 698c576e37894ef7d9cda6d0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75511e209131cea671d6.md>)

Recorded evaluation configuration: {"Pre-trained model":"PointGPT-L","Params. (M)":"3.1","FLOPS (G)":"73.62"}. This record does not assert independent reproduction.

### AsymDSD-B\*

Accuracy: 96.16. Result ID: 698c537737894ef7d9cd8c8e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Implementation](<https://github.com/RFLeijenaar/AsymDSD>)

Recorded evaluation configuration: {"Evaluation Protocol":"Linear","PP (Post-pretraining)":false,"CM (Cross-modal training)":false,"#P(M)":"91.8"}. This record does not assert independent reproduction.

### PointGPT-B\*

Accuracy: 95.8. Result ID: 698c537737894ef7d9cd8c8b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f671e209131cea455b3.md>)

[Implementation](<https://github.com/CGuangyan-BIT/PointGPT>)

Recorded evaluation configuration: {"Evaluation Protocol":"Full Fine-tune","PP (Post-pretraining)":true,"CM (Cross-modal training)":false,"#P(M)":"92.0"}. This record does not assert independent reproduction.

### PointGPT-L

Accuracy: 95.7. Result ID: 698c537737894ef7d9cd8c87.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f671e209131cea455b3.md>)

[Implementation](<https://github.com/CGuangyan-BIT/PointGPT>)

Recorded evaluation configuration: {"Evaluation Protocol":"Full Fine-tune","PP (Post-pretraining)":false,"CM (Cross-modal training)":false,"#P(M)":"310.0"}. This record does not assert independent reproduction.

### Point-RAE

Accuracy: 95.53. Result ID: 698c537837894ef7d9cd8cc2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d949.md>)

Recorded evaluation configuration: {"#P(M)":"29.2","ST (Standard Transformer)":false,"SM (Single-modal training)":true,"Evaluation Protocol":"Full Fine-tune"}. This record does not assert independent reproduction.

### Point-SRA

Accuracy: 95.53. Result ID: 6997eb775698665ec444bbfe.

[Reporting paper](<https://www.sota2.com/research/paper/695c970f53fdd5c361743838.md>)

[Method origin paper](<https://www.sota2.com/research/paper/695c970f53fdd5c361743838.md>)

Recorded evaluation configuration: {"Paradigm":"Cross-Modal Self-Supervised","Params (M)":"40.1"}. This record does not assert independent reproduction.

### AsymDSD-S\*

Accuracy: 95.22. Result ID: 698c537737894ef7d9cd8c8d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Implementation](<https://github.com/RFLeijenaar/AsymDSD>)

Recorded evaluation configuration: {"Evaluation Protocol":"Linear","PP (Post-pretraining)":false,"CM (Cross-modal training)":false,"#P(M)":"21.8"}. This record does not assert independent reproduction.

### Point-FEMAE

Accuracy: 95.18. Result ID: 698c537837894ef7d9cd8cc3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d949.md>)

Recorded evaluation configuration: {"#P(M)":"27.4","ST (Standard Transformer)":false,"SM (Single-modal training)":true,"Evaluation Protocol":"Full Fine-tune"}. This record does not assert independent reproduction.

### ReCon

Accuracy: 95.18. Result ID: 69917ced3a6c314849087c87.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70e91e209131cea4de6b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f721e209131cea45a6b.md>)

[Implementation](<https://github.com/qizekun/ReCon>)

Recorded evaluation configuration: {"#P (M)":"43.6","#F (G)":"5.3","Training Paradigm":"with Pretrained Cross-Modal Teacher Representation Learning"}. This record does not assert independent reproduction.

### Mamba3D

Accuracy: 95.18. Result ID: 6993651e3a6c31484915a11a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ae1e209131cea632e2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74ae1e209131cea632e2.md>)

[Implementation](<https://github.com/xhanxu/Mamba3D>)

Recorded evaluation configuration: {"PT":"Point-MAE","#P (M)":"16.9","#F (G)":"3.9","Voting Strategy":true}. This record does not assert independent reproduction.

### Point-FEMAE

Accuracy: 95.18. Result ID: 6997eb775698665ec444bbf8.

[Reporting paper](<https://www.sota2.com/research/paper/695c970f53fdd5c361743838.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71891e209131cea51d7a.md>)

[Implementation](<https://github.com/zyh16143998882/AAAI24-PointFEMAE>)

Recorded evaluation configuration: {"Paradigm":"Single-Modal Self-Supervised","Params (M)":"41.5"}. This record does not assert independent reproduction.

### ReCon

Accuracy: 95.18. Result ID: 6997eb775698665ec444bbfd.

[Reporting paper](<https://www.sota2.com/research/paper/695c970f53fdd5c361743838.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f721e209131cea45a6b.md>)

[Implementation](<https://github.com/qizekun/ReCon>)

Recorded evaluation configuration: {"Paradigm":"Cross-Modal Self-Supervised","Params (M)":"44.3"}. This record does not assert independent reproduction.

### Utonia

Accuracy: 95. Result ID: 69a8a2946dbb7075b9519f98.

[Reporting paper](<https://www.sota2.com/research/paper/69a7bca02d603e6de3e3d71e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a7bca02d603e6de3e3d71e.md>)

[Implementation](<https://github.com/Pointcept/Utonia>)

Recorded evaluation configuration: {"Params Learn.":"137.4M","Params Pct.":"100%","Evaluation Protocol":"fine-tuning"}. This record does not assert independent reproduction.

### PointGST

Accuracy: 94.49. Result ID: 698c576e37894ef7d9cda6cd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Implementation](<https://github.com/jerryfeng2003/PointGST>)

Recorded evaluation configuration: {"Pre-trained model":"RECON","Params. (M)":"0.6","FLOPS (G)":"4.81"}. This record does not assert independent reproduction.

### Mamba3D

Accuracy: 94.49. Result ID: 6993651e3a6c31484915a10f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ae1e209131cea632e2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74ae1e209131cea632e2.md>)

[Implementation](<https://github.com/xhanxu/Mamba3D>)

Recorded evaluation configuration: {"PT":"×","#P (M)":"16.9","#F (G)":"3.9","Voting Strategy":true}. This record does not assert independent reproduction.

### 3D-JEPA

Accuracy: 94.49. Result ID: 69936a1e3a6c31484915c59a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74e61e209131cea64868.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74e61e209131cea64868.md>)

Recorded evaluation configuration: {"Pre-training Epochs":300,"Training Paradigm":"SSRL"}. This record does not assert independent reproduction.

### AsymDSD-S

Accuracy: 94.32. Result ID: 698c537837894ef7d9cd8cbe.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Implementation](<https://github.com/RFLeijenaar/AsymDSD>)

Recorded evaluation configuration: {"#P(M)":"22.1","ST (Standard Transformer)":true,"SM (Single-modal training)":true,"Evaluation Protocol":"Full Fine-tune"}. This record does not assert independent reproduction.

### PointMamba

Accuracy: 94.32. Result ID: 698c537837894ef7d9cd8cc4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d01e209131cea69fb3.md>)

Recorded evaluation configuration: {"#P(M)":"12.3","ST (Standard Transformer)":false,"SM (Single-modal training)":true,"Evaluation Protocol":"Full Fine-tune"}. This record does not assert independent reproduction.

### Fully fine-tune

Accuracy: 94.32. Result ID: 698c576e37894ef7d9cda6c9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

Recorded evaluation configuration: {"Pre-trained model":"RECON","Params. (M)":"22.1","FLOPS (G)":"4.76"}. This record does not assert independent reproduction.

### DAPT

Accuracy: 94.32. Result ID: 698c576e37894ef7d9cda6cc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72d31e209131cea58c2e.md>)

Recorded evaluation configuration: {"Pre-trained model":"RECON","Params. (M)":"1.1","FLOPS (G)":"4.96"}. This record does not assert independent reproduction.

### PointMamba

Accuracy: 94.32. Result ID: 6997eb775698665ec444bbf6.

[Reporting paper](<https://www.sota2.com/research/paper/695c970f53fdd5c361743838.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d01e209131cea69fb3.md>)

Recorded evaluation configuration: {"Paradigm":"Single-Modal Self-Supervised","Params (M)":"12.3"}. This record does not assert independent reproduction.

### Point-MAE-B\*

Accuracy: 94.2. Result ID: 698c537737894ef7d9cd8c8a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f671e209131cea455b3.md>)

[Implementation](<https://github.com/CGuangyan-BIT/PointGPT>)

Recorded evaluation configuration: {"Evaluation Protocol":"Full Fine-tune","PP (Post-pretraining)":true,"CM (Cross-modal training)":false,"#P(M)":"120.1"}. This record does not assert independent reproduction.

### ReCon SM

Accuracy: 94.15. Result ID: 698c537837894ef7d9cd8cc1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f721e209131cea45a6b.md>)

[Implementation](<https://github.com/qizekun/ReCon>)

Recorded evaluation configuration: {"#P(M)":"43.6","ST (Standard Transformer)":false,"SM (Single-modal training)":true,"Evaluation Protocol":"Full Fine-tune"}. This record does not assert independent reproduction.

### I2P-MAE

Accuracy: 94.15. Result ID: 69917ced3a6c314849087c86.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70e91e209131cea4de6b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e291e209131cea3e3b3.md>)

[Implementation](<https://github.com/ZrrSkywalker/I2P-MAE>)

Recorded evaluation configuration: {"#P (M)":"12.9","#F (G)":"3.6","Training Paradigm":"with Pretrained Cross-Modal Teacher Representation Learning"}. This record does not assert independent reproduction.

### RECON

Accuracy: 94.15. Result ID: 6995a9c13badfaffe6a804e9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78dd1e209131cea7ae14.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f721e209131cea45a6b.md>)

[Implementation](<https://github.com/qizekun/ReCon>)

Recorded evaluation configuration: {"Publication":"ICML 23","Tunable Params.":"43.6 M","Evaluation Protocol":"Self-Supervised Representation Learning (Full Fine-Tuning)"}. This record does not assert independent reproduction.

### I2P-MAE

Accuracy: 94.15. Result ID: 6997eb775698665ec444bbfc.

[Reporting paper](<https://www.sota2.com/research/paper/695c970f53fdd5c361743838.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e291e209131cea3e3b3.md>)

[Implementation](<https://github.com/ZrrSkywalker/I2P-MAE>)

Recorded evaluation configuration: {"Paradigm":"Cross-Modal Self-Supervised","Params (M)":"15.3"}. This record does not assert independent reproduction.

### 3D-JEPA

Accuracy: 93.8. Result ID: 69936a1e3a6c31484915c599.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74e61e209131cea64868.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74e61e209131cea64868.md>)

Recorded evaluation configuration: {"Pre-training Epochs":150,"Training Paradigm":"SSRL"}. This record does not assert independent reproduction.

### PointNTP

Accuracy: 93.8. Result ID: 6a0c419bfbc373592c0324b3.

[Reporting paper](<https://www.sota2.com/research/paper/6a0beee6dbc836b87d82c7f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0beee6dbc836b87d82c7f0.md>)

Recorded evaluation configuration: {"Pretrain Framework":"fully causal autoregression","Auxiliary Module":"none"}. This record does not assert independent reproduction.

### Point-MAE† w/ IDPT

Accuracy: 93.63. Result ID: 699120b13a6c31484905e653.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70231e209131cea49508.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70231e209131cea49508.md>)

[Implementation](<https://github.com/zyh16143998882/ICCV23-IDPT>)

Recorded evaluation configuration: {"#TP (M)":"1.7","Learning Paradigm":"Self-Supervised Representation Learning (IDPT)"}. This record does not assert independent reproduction.

### PointGPT-B

Accuracy: 93.6. Result ID: 698c537737894ef7d9cd8c86.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f671e209131cea455b3.md>)

[Implementation](<https://github.com/CGuangyan-BIT/PointGPT>)

Recorded evaluation configuration: {"Evaluation Protocol":"Full Fine-tune","PP (Post-pretraining)":false,"CM (Cross-modal training)":false,"#P(M)":"92.0"}. This record does not assert independent reproduction.

### PointGST

Accuracy: 93.46. Result ID: 698c576e37894ef7d9cda6c8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Implementation](<https://github.com/jerryfeng2003/PointGST>)

Recorded evaluation configuration: {"Pre-trained model":"ACT","Params. (M)":"0.6","FLOPS (G)":"4.81"}. This record does not assert independent reproduction.

### Standard Transformer

Accuracy: 93.29. Result ID: 698b4af78d67a41d0a6cdcbf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79651e209131cea7d694.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79651e209131cea7d694.md>)

[Implementation](<https://github.com/wangzy22/UniPre3D>)

Recorded evaluation configuration: {"Pre-train":"PointDif \[82\]"}. This record does not assert independent reproduction.

### Fully fine-tune

Accuracy: 93.29. Result ID: 698c576e37894ef7d9cda6c4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

Recorded evaluation configuration: {"Pre-trained model":"ACT","Params. (M)":"22.1","FLOPS (G)":"4.76"}. This record does not assert independent reproduction.

### IDPT

Accuracy: 93.29. Result ID: 698c576e37894ef7d9cda6ca.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70231e209131cea49508.md>)

[Implementation](<https://github.com/zyh16143998882/ICCV23-IDPT>)

Recorded evaluation configuration: {"Pre-trained model":"RECON","Params. (M)":"1.7","FLOPS (G)":"7.10"}. This record does not assert independent reproduction.

### ACT

Accuracy: 93.29. Result ID: 699120b13a6c31484905e64e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70231e209131cea49508.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e741e209131cea3ff1b.md>)

[Implementation](<https://github.com/RunpeiDong/ACT>)

Recorded evaluation configuration: {"#TP (M)":"22.1","Learning Paradigm":"Self-Supervised Representation Learning (Full Fine-tuning)"}. This record does not assert independent reproduction.

### ACT

Accuracy: 93.29. Result ID: 69917ced3a6c314849087c85.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70e91e209131cea4de6b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e741e209131cea3ff1b.md>)

[Implementation](<https://github.com/RunpeiDong/ACT>)

Recorded evaluation configuration: {"#P (M)":"22.1","#F (G)":"4.8","Training Paradigm":"with Pretrained Cross-Modal Teacher Representation Learning"}. This record does not assert independent reproduction.

### ACT

Accuracy: 93.29. Result ID: 69936a1e3a6c31484915c596.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74e61e209131cea64868.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e741e209131cea3ff1b.md>)

[Implementation](<https://github.com/RunpeiDong/ACT>)

Recorded evaluation configuration: {"Pre-training Epochs":300,"Training Paradigm":"SSRL"}. This record does not assert independent reproduction.

### ACT

Accuracy: 93.29. Result ID: 6995a9c13badfaffe6a804e8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78dd1e209131cea7ae14.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e741e209131cea3ff1b.md>)

[Implementation](<https://github.com/RunpeiDong/ACT>)

Recorded evaluation configuration: {"Publication":"ICLR'23","Tunable Params.":"22.1 M","Evaluation Protocol":"Self-Supervised Representation Learning (Full Fine-Tuning)"}. This record does not assert independent reproduction.

### PointDif

Accuracy: 93.29. Result ID: 6997eb775698665ec444bbf7.

[Reporting paper](<https://www.sota2.com/research/paper/695c970f53fdd5c361743838.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa714a1e209131cea50790.md>)

Recorded evaluation configuration: {"Paradigm":"Single-Modal Self-Supervised"}. This record does not assert independent reproduction.

### ACT

Accuracy: 93.29. Result ID: 6997eb775698665ec444bbfb.

[Reporting paper](<https://www.sota2.com/research/paper/695c970f53fdd5c361743838.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e741e209131cea3ff1b.md>)

[Implementation](<https://github.com/RunpeiDong/ACT>)

Recorded evaluation configuration: {"Paradigm":"Cross-Modal Self-Supervised","Params (M)":"22.1"}. This record does not assert independent reproduction.

### PointDif

Accuracy: 93.29. Result ID: 6a0c419bfbc373592c0324b1.

[Reporting paper](<https://www.sota2.com/research/paper/6a0beee6dbc836b87d82c7f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa714a1e209131cea50790.md>)

Recorded evaluation configuration: {"Pretrain Framework":"diffusion-based pretraining","Auxiliary Module":"conditional point generator"}. This record does not assert independent reproduction.

### ACT

Accuracy: 93.29. Result ID: 6a0c419bfbc373592c0324b6.

[Reporting paper](<https://www.sota2.com/research/paper/6a0beee6dbc836b87d82c7f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e741e209131cea3ff1b.md>)

[Implementation](<https://github.com/RunpeiDong/ACT>)

Recorded evaluation configuration: {"Pretrain Framework":"cross-modal teacher-guided masked modeling","Auxiliary Module":"teacher autoencoder + transformer decoder"}. This record does not assert independent reproduction.

### Point-JEPA

Accuracy: 93.2. Result ID: 6997eb775698665ec444bbf9.

[Reporting paper](<https://www.sota2.com/research/paper/695c970f53fdd5c361743838.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76ea1e209131cea6fea8.md>)

[Implementation](<https://github.com/Ayumu-J-S/Point-JEPA>)

Recorded evaluation configuration: {"Paradigm":"Single-Modal Self-Supervised"}. This record does not assert independent reproduction.

### IDPT

Accuracy: 93.12. Result ID: 698c576e37894ef7d9cda6c5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70231e209131cea49508.md>)

[Implementation](<https://github.com/zyh16143998882/ICCV23-IDPT>)

Recorded evaluation configuration: {"Pre-trained model":"ACT","Params. (M)":"1.7","FLOPS (G)":"7.10"}. This record does not assert independent reproduction.

### ACT w/ IDPT

Accuracy: 93.12. Result ID: 699120b13a6c31484905e652.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70231e209131cea49508.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70231e209131cea49508.md>)

[Implementation](<https://github.com/zyh16143998882/ICCV23-IDPT>)

Recorded evaluation configuration: {"#TP (M)":"1.7","Learning Paradigm":"Self-Supervised Representation Learning (IDPT)"}. This record does not assert independent reproduction.

### Mamba3D

Accuracy: 93.12. Result ID: 6993651e3a6c31484915a119.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ae1e209131cea632e2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74ae1e209131cea632e2.md>)

[Implementation](<https://github.com/xhanxu/Mamba3D>)

Recorded evaluation configuration: {"PT":"Point-MAE","#P (M)":"16.9","#F (G)":"3.9","Voting Strategy":false}. This record does not assert independent reproduction.

### VPP w/ vot.

Accuracy: 93.11. Result ID: 69917ced3a6c314849087c84.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70e91e209131cea4de6b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70e91e209131cea4de6b.md>)

[Implementation](<https://github.com/qizekun/VPP>)

Recorded evaluation configuration: {"#P (M)":"22.1","#F (G)":"4.8","Training Paradigm":"with Self-Supervised Representation Learning","Voting Strategy":"Yes"}. This record does not assert independent reproduction.

### Point-MAE†

Accuracy: 92.94. Result ID: 699120b13a6c31484905e64f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70231e209131cea49508.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c1b1e209131cea32397.md>)

[Implementation](<https://github.com/Pang-Yatian/Point-MAE>)

Recorded evaluation configuration: {"#TP (M)":"22.1","Learning Paradigm":"Self-Supervised Representation Learning (Full Fine-tuning)"}. This record does not assert independent reproduction.

### Mamba3D

Accuracy: 92.94. Result ID: 6993651e3a6c31484915a10e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ae1e209131cea632e2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74ae1e209131cea632e2.md>)

[Implementation](<https://github.com/xhanxu/Mamba3D>)

Recorded evaluation configuration: {"PT":"×","#P (M)":"16.9","#F (G)":"3.9","Voting Strategy":false}. This record does not assert independent reproduction.

### AsymSD-CLS-S

Accuracy: 92.77. Result ID: 698c537837894ef7d9cd8cbc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Implementation](<https://github.com/RFLeijenaar/AsymDSD>)

Recorded evaluation configuration: {"#P(M)":"22.1","ST (Standard Transformer)":true,"SM (Single-modal training)":true,"Evaluation Protocol":"Full Fine-tune"}. This record does not assert independent reproduction.

### AsymSD-MPM-S

Accuracy: 92.77. Result ID: 698c537837894ef7d9cd8cbd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Implementation](<https://github.com/RFLeijenaar/AsymDSD>)

Recorded evaluation configuration: {"#P(M)":"22.1","ST (Standard Transformer)":true,"SM (Single-modal training)":true,"Evaluation Protocol":"Full Fine-tune"}. This record does not assert independent reproduction.

### VPP w/o vot.

Accuracy: 92.77. Result ID: 69917ced3a6c314849087c83.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70e91e209131cea4de6b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70e91e209131cea4de6b.md>)

[Implementation](<https://github.com/qizekun/VPP>)

Recorded evaluation configuration: {"#P (M)":"22.1","#F (G)":"4.8","Training Paradigm":"with Self-Supervised Representation Learning","Voting Strategy":"No"}. This record does not assert independent reproduction.

### Standard Transformer

Accuracy: 92.6. Result ID: 698b4af78d67a41d0a6cdcc0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79651e209131cea7d694.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79651e209131cea7d694.md>)

[Implementation](<https://github.com/wangzy22/UniPre3D>)

Recorded evaluation configuration: {"Pre-train":"UniPre3D"}. This record does not assert independent reproduction.

### DAPT

Accuracy: 92.6. Result ID: 698c576e37894ef7d9cda6c7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72d31e209131cea58c2e.md>)

Recorded evaluation configuration: {"Pre-trained model":"ACT","Params. (M)":"1.1","FLOPS (G)":"4.96"}. This record does not assert independent reproduction.

### MVTN

Accuracy: 92.6. Result ID: 6993651e3a6c31484915a10b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ae1e209131cea632e2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b40.md>)

[Implementation](<https://github.com/ajhamdi/MVTN>)

Recorded evaluation configuration: {"PT":"×","#P (M)":"11.2","#F (G)":"43.7"}. This record does not assert independent reproduction.

### MVTN

Accuracy: 92.6. Result ID: 6a0c419bfbc373592c0324a8.

[Reporting paper](<https://www.sota2.com/research/paper/6a0beee6dbc836b87d82c7f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b40.md>)

[Implementation](<https://github.com/ajhamdi/MVTN>)

Recorded evaluation configuration: {"Pretrain Framework":"multi-view supervised learning","Auxiliary Module":"N/A"}. This record does not assert independent reproduction.

### UniPre3D (Std. Transformer)

Accuracy: 92.6. Result ID: 6a0c419bfbc373592c0324b7.

[Reporting paper](<https://www.sota2.com/research/paper/6a0beee6dbc836b87d82c7f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79651e209131cea7d694.md>)

[Implementation](<https://github.com/wangzy22/UniPre3D>)

Recorded evaluation configuration: {"Pretrain Framework":"cross-modal Gaussian splatting pretraining","Auxiliary Module":"Gaussian predictor + fusion block"}. This record does not assert independent reproduction.

### IAE (M2AE)

Accuracy: 92.5. Result ID: 699ab2c95698665ec4405fa9.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad1b6cfd488761a4e8a5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6937ad1b6cfd488761a4e8a5.md>)

Recorded evaluation configuration: {"#Params(M)":"15.3","GFLOPS":"3.6","Evaluation Protocol":"FULL SSL"}. This record does not assert independent reproduction.

### IAE (M2AE)

Accuracy: 92.5. Result ID: 6a0c419bfbc373592c0324af.

[Reporting paper](<https://www.sota2.com/research/paper/6a0beee6dbc836b87d82c7f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70771e209131cea4b43e.md>)

[Implementation](<https://github.com/SimingYan/IAE>)

Recorded evaluation configuration: {"Pretrain Framework":"implicit autoencoder","Auxiliary Module":"implicit decoder"}. This record does not assert independent reproduction.

### Concerto

Accuracy: 92.3. Result ID: 69a8a2946dbb7075b9519f97.

[Reporting paper](<https://www.sota2.com/research/paper/69a7bca02d603e6de3e3d71e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69004e163ef7ccd74e5e2ee3.md>)

Recorded evaluation configuration: {"Params Learn.":"137.4M","Params Pct.":"100%","Evaluation Protocol":"fine-tuning"}. This record does not assert independent reproduction.

### IAE (M2AE)

Accuracy: 92.3. Result ID: 699ab2c95698665ec4405fa8.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad1b6cfd488761a4e8a5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6937ad1b6cfd488761a4e8a5.md>)

Recorded evaluation configuration: {"#Params(M)":"15.3","GFLOPS":"3.6","Evaluation Protocol":"FULL SSL","pre-training mesh":"w/o mesh"}. This record does not assert independent reproduction.

### Mamba3D

Accuracy: 92.25. Result ID: 6993651e3a6c31484915a117.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ae1e209131cea632e2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74ae1e209131cea632e2.md>)

[Implementation](<https://github.com/xhanxu/Mamba3D>)

Recorded evaluation configuration: {"PT":"Point-BERT","#P (M)":"16.9","#F (G)":"3.9","Voting Strategy":false}. This record does not assert independent reproduction.

### Sonata

Accuracy: 92.2. Result ID: 69a8a2946dbb7075b9519f96.

[Reporting paper](<https://www.sota2.com/research/paper/69a7bca02d603e6de3e3d71e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77811e209131cea73a63.md>)

[Implementation](<https://github.com/facebookresearch/sonata>)

Recorded evaluation configuration: {"Params Learn.":"124.8M","Params Pct.":"100%","Evaluation Protocol":"fine-tuning"}. This record does not assert independent reproduction.

### Point-PEFT

Accuracy: 91.91. Result ID: 698c576e37894ef7d9cda6cb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75511e209131cea671d6.md>)

Recorded evaluation configuration: {"Pre-trained model":"RECON","Params. (M)":"0.7","FLOPS (G)":"7.61"}. This record does not assert independent reproduction.

### RI-MAE

Accuracy: 91.9. Result ID: 6997eb775698665ec444bbfa.

[Reporting paper](<https://www.sota2.com/research/paper/695c970f53fdd5c361743838.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76581e209131cea6c830.md>)

Recorded evaluation configuration: {"Paradigm":"Single-Modal Self-Supervised"}. This record does not assert independent reproduction.

### PointGST

Accuracy: 91.74. Result ID: 698c576e37894ef7d9cda6c3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Implementation](<https://github.com/jerryfeng2003/PointGST>)

Recorded evaluation configuration: {"Pre-trained model":"Point-MAE","Params. (M)":"0.6","FLOPS (G)":"4.81"}. This record does not assert independent reproduction.

### PointGPT-S

Accuracy: 91.6. Result ID: 698c537837894ef7d9cd8cbb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f671e209131cea455b3.md>)

[Implementation](<https://github.com/CGuangyan-BIT/PointGPT>)

Recorded evaluation configuration: {"#P(M)":"22.1","ST (Standard Transformer)":true,"SM (Single-modal training)":true,"Evaluation Protocol":"Full Fine-tune"}. This record does not assert independent reproduction.

### PointGPT-S

Accuracy: 91.6. Result ID: 6a0c419bfbc373592c0324b0.

[Reporting paper](<https://www.sota2.com/research/paper/6a0beee6dbc836b87d82c7f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f671e209131cea455b3.md>)

[Implementation](<https://github.com/CGuangyan-BIT/PointGPT>)

Recorded evaluation configuration: {"Pretrain Framework":"autoregressive generation","Auxiliary Module":"extractor-generator transformer decoder"}. This record does not assert independent reproduction.

### PointGST

Accuracy: 91.39. Result ID: 698c576e37894ef7d9cda6be.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Implementation](<https://github.com/jerryfeng2003/PointGST>)

Recorded evaluation configuration: {"Pre-trained model":"Point-BERT","Params. (M)":"0.6","FLOPS (G)":"4.81"}. This record does not assert independent reproduction.

### Point-M2AE

Accuracy: 91.22. Result ID: 698c537837894ef7d9cd8cc0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79b41e209131cea7efc4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6daa1e209131cea3b524.md>)

[Implementation](<https://github.com/ZrrSkywalker/Point-M2AE>)

Recorded evaluation configuration: {"#P(M)":"15.3","ST (Standard Transformer)":false,"SM (Single-modal training)":true,"Evaluation Protocol":"Full Fine-tune"}. This record does not assert independent reproduction.

### IDPT

Accuracy: 91.22. Result ID: 698c576e37894ef7d9cda6c0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a3a1e209131cea81f2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70231e209131cea49508.md>)

[Implementation](<https://github.com/zyh16143998882/ICCV23-IDPT>)

Recorded evaluation configuration: {"Pre-trained model":"Point-MAE","Params. (M)":"1.7","FLOPS (G)":"7.10"}. This record does not assert independent reproduction.

### Point-M2AE

Accuracy: 91.22. Result ID: 699120b13a6c31484905e64d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70231e209131cea49508.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6daa1e209131cea3b524.md>)

[Implementation](<https://github.com/ZrrSkywalker/Point-M2AE>)

Recorded evaluation configuration: {"Learning Paradigm":"Self-Supervised Representation Learning (Full Fine-tuning)"}. This record does not assert independent reproduction.

### Point-MAE w/ IDPT

Accuracy: 91.22. Result ID: 699120b13a6c31484905e651.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70231e209131cea49508.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70231e209131cea49508.md>)

[Implementation](<https://github.com/zyh16143998882/ICCV23-IDPT>)

Recorded evaluation configuration: {"#TP (M)":"1.7","Learning Paradigm":"Self-Supervised Representation Learning (IDPT)"}. This record does not assert independent reproduction.

Page 1 of 4. This is an index page, not the full collection.

- [Next index page](<https://www.sota2.com/research/sota/object-classification-on-scanobjectnn-obj_bg.md?metric=Accuracy&page=2>): Follow this link to discover more pages.
