3D Object Recognition on ModelNet40
Accuracy
- 2016-04-12 — MVCNN-MultiRes: Accuracy 93.8
- 2021-10-25 — MVT-small: Accuracy 97.5
- 2022-11-20 — R2-MLP-36: Accuracy 97.7
| Rank | Model | Accuracy | Paper | Code | Year |
|---|---|---|---|---|---|
| 1 | R2-MLP-36 | 97.7% | R2-MLP: Round-Roll MLP for Multi-View 3D Object Recognition | shanshuo/MVT · shanshuo/R2-MLP | 2022 |
| 2 | MVT-small | 97.5% | MVT: Multi-view Vision Transformer for 3D Object Recognition | shanshuo/MVT · shanshuo/R2-MLP | 2021 |
| 3 | MVCNN-MultiRes | 93.8% | Volumetric and Multi-View CNNs for Object Classification on 3D Data | charlesq34/3dcnn.torch · LONG-9621/3DCNN | 2016 |
| 4 | MeshWalker (ours) | 92.3% | MeshWalker: Deep Mesh Understanding by Random Walks | AlonLahav/MeshWalker | 2020 |
| 5 | FPNN (4-FCs + NF) | 88.4% | FPNN: Field Probing Neural Networks for 3D Data | yangyanli/FPNN · MindSpore-scientific-2/code-3 | 2016 |
| 6 | Variational Shape Learner | 84.5% | Learning a Hierarchical Latent-Variable Model of 3D Shapes | lorenmt/vsl | 2017 |