# 3D Point Cloud Classification on ShapeNet-C Source: [3D Point Cloud Classification on ShapeNet-C]() Collection index: [llms.txt]() Data updated: 2026-06-12T07:28:16.771Z Evaluation of 3D Point Cloud Classification performance on the ShapeNet-C dataset, reporting robustness metrics across various corruption types (e.g., uniform noise, Gaussian noise, background addition, impulse noise, etc.). 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 | 3D Point Cloud Classification | | Dataset (including any stated split) | ShapeNet-C | | Metric | Accuracy (Uniform Noise) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 17 of 17 | | 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 (Uniform Noise)](): Higher is better; unit: PERCENT. - [Accuracy (Gaussian Noise)](): Higher is better; unit: PERCENT. - [Accuracy (Background Addition)](): Higher is better; unit: PERCENT. - [Accuracy (Impulse Noise)](): Higher is better; unit: PERCENT. - [Accuracy (Upsampling)](): Higher is better; unit: PERCENT. - [Accuracy (RBF)](): Higher is better; unit: PERCENT. - [Accuracy (RBF Inverse)](): Higher is better; unit: PERCENT. - [Accuracy (Density Decrease)](): Higher is better; unit: PERCENT. - [Accuracy (Density Increase)](): Higher is better; unit: PERCENT. - [Accuracy (Shear)](): Higher is better; unit: PERCENT. - [Accuracy (Rotation)](): Higher is better; unit: PERCENT. - [Accuracy (Cutout)](): Higher is better; unit: PERCENT. - [Accuracy (Distortion)](): Higher is better; unit: PERCENT. - [Accuracy (Occlusion)](): Higher is better; unit: PERCENT. - [Accuracy (LiDAR)](): Higher is better; unit: PERCENT. - [Mean Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (Uniform Noise) | Result date | Report paper | | --- | --- | --- | --- | | SVWA | 82.04 | Not recorded | 6a2b92bad33dfb23e783b176 | | MAMVI | 81.7 | Not recorded | 6a2b92bad33dfb23e783b176 | | SMART-PC | 80.8 | Not recorded | 6a2b92bad33dfb23e783b176 | | PG-SF | 79.21 | Not recorded | 6a2b92bad33dfb23e783b176 | | PG-SP | 78.63 | Not recorded | 6a2b92bad33dfb23e783b176 | | MATE | 77.8 | Not recorded | 6a2b92bad33dfb23e783b176 | | Source Only | 77.34 | Not recorded | 6a2b92bad33dfb23e783b176 | | DUA | 76.1 | Not recorded | 6a2b92bad33dfb23e783b176 | | MEMO | 75.27 | Not recorded | 6a2b92bad33dfb23e783b176 | | TTT-Rot | 74.6 | Not recorded | 6a2b92bad33dfb23e783b176 | | LAME | 72.88 | Not recorded | 6a2b92bad33dfb23e783b176 | | BFTT3D | 72.58 | Not recorded | 6a2b92bad33dfb23e783b176 | | SAR | 72.49 | Not recorded | 6a2b92bad33dfb23e783b176 | | TENT | 71.9 | Not recorded | 6a2b92bad33dfb23e783b176 | | PL | 70.3 | Not recorded | 6a2b92bad33dfb23e783b176 | | T3A | 66.69 | Not recorded | 6a2b92bad33dfb23e783b176 | | SHOT | 27.78 | Not recorded | 6a2b92bad33dfb23e783b176 | ### SVWA Accuracy (Uniform Noise): 82.04. Result ID: 6a2bb504fbc373592c29c029. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Adaptation Protocol":"Source-Free Methods"}. This record does not assert independent reproduction. ### MAMVI Accuracy (Uniform Noise): 81.7. Result ID: 6a2bb504fbc373592c29c02a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Adaptation Protocol":"Source-Free Methods"}. This record does not assert independent reproduction. ### SMART-PC Accuracy (Uniform Noise): 80.8. Result ID: 6a2bb504fbc373592c29c01d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Adaptation Protocol":"TTT Methods"}. This record does not assert independent reproduction. ### PG-SF Accuracy (Uniform Noise): 79.21. Result ID: 6a2bb504fbc373592c29c028. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Adaptation Protocol":"Source-Free Methods"}. This record does not assert independent reproduction. ### PG-SP Accuracy (Uniform Noise): 78.63. Result ID: 6a2bb504fbc373592c29c01f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Adaptation Protocol":"Source-Dependent Methods"}. This record does not assert independent reproduction. ### MATE Accuracy (Uniform Noise): 77.8. Result ID: 6a2bb504fbc373592c29c01c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Adaptation Protocol":"TTT Methods"}. This record does not assert independent reproduction. ### Source Only Accuracy (Uniform Noise): 77.34. Result ID: 6a2bb504fbc373592c29c01a. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DUA Accuracy (Uniform Noise): 76.1. Result ID: 6a2bb504fbc373592c29c025. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Adaptation Protocol":"Source-Free Methods"}. This record does not assert independent reproduction. ### MEMO Accuracy (Uniform Noise): 75.27. Result ID: 6a2bb504fbc373592c29c026. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Adaptation Protocol":"Source-Free Methods"}. This record does not assert independent reproduction. ### TTT-Rot Accuracy (Uniform Noise): 74.6. Result ID: 6a2bb504fbc373592c29c01b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Adaptation Protocol":"TTT Methods"}. This record does not assert independent reproduction. ### LAME Accuracy (Uniform Noise): 72.88. Result ID: 6a2bb504fbc373592c29c024. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Adaptation Protocol":"Source-Free Methods"}. This record does not assert independent reproduction. ### BFTT3D Accuracy (Uniform Noise): 72.58. Result ID: 6a2bb504fbc373592c29c01e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Adaptation Protocol":"Source-Dependent Methods"}. This record does not assert independent reproduction. ### SAR Accuracy (Uniform Noise): 72.49. Result ID: 6a2bb504fbc373592c29c027. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Adaptation Protocol":"Source-Free Methods"}. This record does not assert independent reproduction. ### TENT Accuracy (Uniform Noise): 71.9. Result ID: 6a2bb504fbc373592c29c021. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Adaptation Protocol":"Source-Free Methods"}. This record does not assert independent reproduction. ### PL Accuracy (Uniform Noise): 70.3. Result ID: 6a2bb504fbc373592c29c020. [Reporting paper]() Recorded evaluation configuration: {"Adaptation Protocol":"Source-Free Methods"}. This record does not assert independent reproduction. ### T3A Accuracy (Uniform Noise): 66.69. Result ID: 6a2bb504fbc373592c29c023. [Reporting paper]() Recorded evaluation configuration: {"Adaptation Protocol":"Source-Free Methods"}. This record does not assert independent reproduction. ### SHOT Accuracy (Uniform Noise): 27.78. Result ID: 6a2bb504fbc373592c29c022. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Adaptation Protocol":"Source-Free Methods"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.