{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/3d-lidar-and-stereo-fusion-using-stereo","title":"3D LiDAR and Stereo Fusion using Stereo Matching Network with Conditional Cost Volume Normalization","arxiv_id":"1904.02917","date":"2019-04-05","proceeding":null,"authors":["Tsun-Hsuan Wang","Hou-Ning Hu","Chieh Hubert Lin","Yi-Hsuan Tsai","Wei-Chen Chiu","Min Sun"],"abstract":"The complementary characteristics of active and passive depth sensing\ntechniques motivate the fusion of the Li-DAR sensor and stereo camera for\nimproved depth perception. Instead of directly fusing estimated depths across\nLiDAR and stereo modalities, we take advantages of the stereo matching network\nwith two enhanced techniques: Input Fusion and Conditional Cost Volume\nNormalization (CCVNorm) on the LiDAR information. The proposed framework is\ngeneric and closely integrated with the cost volume component that is commonly\nutilized in stereo matching neural networks. We experimentally verify the\nefficacy and robustness of our method on the KITTI Stereo and Depth Completion\ndatasets, obtaining favorable performance against various fusion strategies.\nMoreover, we demonstrate that, with a hierarchical extension of CCVNorm, the\nproposed method brings only slight overhead to the stereo matching network in\nterms of computation time and model size. For project page, see\nhttps://zswang666.github.io/Stereo-LiDAR-CCVNorm-Project-Page/","url_abs":"http://arxiv.org/abs/1904.02917v1","url_pdf":"http://arxiv.org/pdf/1904.02917v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"3d-lidar-and-stereo-fusion-using-stereo","repo_url":"https://github.com/zswang666/Stereo-LiDAR-CCVNorm","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"depth-completion","task_name":"Depth Completion"},{"task_slug":"stereo-matching-1","task_name":"Stereo Matching"},{"task_slug":"stereo-matching","task_name":"Stereo Matching Hand"},{"task_slug":"stereo-lidar-fusion","task_name":"Stereo-LiDAR Fusion"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/stereo-lidar-fusion-on-kitti-depth-completion","task":"Stereo-LiDAR Fusion","dataset":"KITTI Depth Completion Validation","model":"CCVN","rank_in_archive_order":3,"of":9,"metrics":{"RMSE":"749.3"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1904.02917","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}