---
license: mit
tags:
- 3d-object-detection
- point-cloud
- scannet
- sun-rgbd
---

# Canonical Voting: Pretrained Models

Pretrained models for [Canonical Voting: Towards Robust Oriented Bounding Box Detection in 3D Scenes](https://openaccess.thecvf.com/content/CVPR2022/html/You_Canonical_Voting_Towards_Robust_Oriented_Bounding_Box_Detection_in_3D_CVPR_2022_paper.html) (CVPR 2022).

Code: https://github.com/qq456cvb/CanonicalVoting

## Files

| File | Description |
| --- | --- |
| `scannet/joint.pth` | Jointly trained model for all categories on ScanNet (~15.4 mAP) |
| `scannet/separate/<wordnet_id>.pth` | Separately trained per-category models on ScanNet (~21.7 mAP overall) |
| `sunrgbd/checkpoint.pth` | Pretrained CanonicalVoting model for SUN RGB-D (used with BRNetCanon) |

Per-category ScanNet models cover wordnet ids `02747177`, `02808440`, `02871439`, `02933112`, `03001627`, `03211117`, `04256520`, `04379243`, plus `others`.

## Citation

```
@inproceedings{you2022canonical,
  title={Canonical Voting: Towards Robust Oriented Bounding Box Detection in 3D Scenes},
  author={You, Yang and Ye, Zelin and Lou, Yujing and Li, Chengkun and Li, Yong-Lu and Ma, Lizhuang and Wang, Weiming and Lu, Cewu},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  pages={1193--1202},
  year={2022}
}
```