Noisy, partially overlapping data and the need for real-time processing pose major challenges for rigid registration. Considering that feature-based matching can handle large transformation differences but suffers from limited accuracy, while local geometry-based matching can achieve fine-grained local alignment but relies heavily on a good initial transformation, we propose a novel dual-space paradigm to fully leverage the strengths of both approaches. First, we introduce an efficient filtering mechanism consisting of a computationally lightweight one-point RANSAC algorithm and a subsequent refinement module to eliminate unreliable feature-based correspondences. Subsequently, we treat the filtered correspondences as anchor points, extract geometric proxies, and formulate an effective objective function with a tailored solver to estimate the transformation. Experiments verify our method's effectiveness, as demonstrated by a 32x CPU-time speedup over MAC on KITTI with comparable accuracy. Project page: https://ustc3dv.github.io/DualReg/.
@article{arxiv.2508.17034,
title = {DualReg: Dual-Space Filtering and Reinforcement for Rigid Registration},
author = {Jiayi Li and Yuxin Yao and Qiuhang Lu and Juyong Zhang},
journal= {arXiv preprint arXiv:2508.17034},
year = {2026}
}
Comments
Accepted to CVPR 2026, Project page: https://ustc3dv.github.io/DualReg/