Point cloud registration is a task to estimate the rigid transformation between two unaligned scans, which plays an important role in many computer vision applications. Previous learning-based works commonly focus on supervised registration, which have limitations in practice. Recently, with the advance of inexpensive RGB-D sensors, several learning-based works utilize RGB-D data to achieve unsupervised registration. However, most of existing unsupervised methods follow a cascaded design or fuse RGB-D data in a unidirectional manner, which do not fully exploit the complementary information in the RGB-D data. To leverage the complementary information more effectively, we propose a network implementing multi-scale bidirectional fusion between RGB images and point clouds generated from depth images. By bidirectionally fusing visual and geometric features in multi-scales, more distinctive deep features for correspondence estimation can be obtained, making our registration more accurate. Extensive experiments on ScanNet and 3DMatch demonstrate that our method achieves new state-of-the-art performance. Code will be released at https://github.com/phdymz/PointMBF
@article{arxiv.2308.04782,
title = {PointMBF: A Multi-scale Bidirectional Fusion Network for Unsupervised RGB-D Point Cloud Registration},
author = {Mingzhi Yuan and Kexue Fu and Zhihao Li and Yucong Meng and Manning Wang},
journal= {arXiv preprint arXiv:2308.04782},
year = {2023}
}