{"ID":2824352,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.23472","arxiv_id":"2512.23472","title":"MCI-Net: A Robust Multi-Domain Context Integration Network for Point Cloud Registration","abstract":"Robust and discriminative feature learning is critical for high-quality point cloud registration. However, existing deep learning-based methods typically rely on Euclidean neighborhood-based strategies for feature extraction, which struggle to effectively capture the implicit semantics and structural consistency in point clouds. To address these issues, we propose a multi-domain context integration network (MCI-Net) that improves feature representation and registration performance by aggregating contextual cues from diverse domains. Specifically, we propose a graph neighborhood aggregation module, which constructs a global graph to capture the overall structural relationships within point clouds. We then propose a progressive context interaction module to enhance feature discriminability by performing intra-domain feature decoupling and inter-domain context interaction. Finally, we design a dynamic inlier selection method that optimizes inlier weights using residual information from multiple iterations of pose estimation, thereby improving the accuracy and robustness of registration. Extensive experiments on indoor RGB-D and outdoor LiDAR datasets show that the proposed MCI-Net significantly outperforms existing state-of-the-art methods, achieving the highest registration recall of 96.4\\% on 3DMatch. Source code is available at http://www.linshuyuan.com.","short_abstract":"Robust and discriminative feature learning is critical for high-quality point cloud registration. However, existing deep learning-based methods typically rely on Euclidean neighborhood-based strategies for feature extraction, which struggle to effectively capture the implicit semantics and structural consistency in poi...","url_abs":"https://arxiv.org/abs/2512.23472","url_pdf":"https://arxiv.org/pdf/2512.23472v1","authors":"[\"Shuyuan Lin\",\"Wenwu Peng\",\"Junjie Huang\",\"Qiang Qi\",\"Miaohui Wang\",\"Jian Weng\"]","published":"2025-12-29T13:55:33Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","project_urls":"[\"http://www.linshuyuan.com\"]","has_code":false,"code_links":[{"ID":605575,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2824352,"paper_url":"https://arxiv.org/abs/2512.23472","paper_title":"MCI-Net: A Robust Multi-Domain Context Integration Network for Point Cloud Registration","repo_url":"https://github.com/shuyuanlin/SCNet","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0},{"ID":605576,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2824352,"paper_url":"https://arxiv.org/abs/2512.23472","paper_title":"MCI-Net: A Robust Multi-Domain Context Integration Network for Point Cloud Registration","repo_url":"https://github.com/shuyuanlin/MCINet","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0},{"ID":605577,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2824352,"paper_url":"https://arxiv.org/abs/2512.23472","paper_title":"MCI-Net: A Robust Multi-Domain Context Integration Network for Point Cloud Registration","repo_url":"https://github.com/shuyuanlin/LLHANet","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0},{"ID":605578,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2824352,"paper_url":"https://arxiv.org/abs/2512.23472","paper_title":"MCI-Net: A Robust Multi-Domain Context Integration Network for Point Cloud Registration","repo_url":"https://github.com/shuyuanlin/MGCANET","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0},{"ID":605579,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2824352,"paper_url":"https://arxiv.org/abs/2512.23472","paper_title":"MCI-Net: A Robust Multi-Domain Context Integration Network for Point Cloud Registration","repo_url":"https://github.com/shuyuanlin/MSGSA","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0},{"ID":605580,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2824352,"paper_url":"https://arxiv.org/abs/2512.23472","paper_title":"MCI-Net: A Robust Multi-Domain Context Integration Network for Point Cloud Registration","repo_url":"https://github.com/shuyuanlin/DTSNET","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}