TEAR-release

Implementation of the paper:

[Scalable 3D Registration via Truncated Entry-wise Absolute Residuals] (CVPR 2024).

To apply this algorithm to your data, you may need to adjust the noise_bound and BRANCH_ACCURACY. For example, smaller BRANCH_ACCURACY leads to higher accuracy but requires more time cost.

Citation

@inproceedings{huang2024scalable,
  title={Scalable 3d registration via truncated entry-wise absolute residuals},
  author={Huang, Tianyu and Peng, Liangzu and Vidal, Ren{\'e} and Liu, Yun-Hui},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages={27477--27487},
  year={2024}
}

Our another work regarding 3D registration

Efficient and Robust Point Cloud Registration via Heuristics-guided Parameter Search (T-PAMI 2024)