{"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/deepmapping-unsupervised-map-estimation-from","title":"DeepMapping: Unsupervised Map Estimation From Multiple Point Clouds","arxiv_id":"1811.11397","date":"2018-11-28","proceeding":"CVPR 2019 6","authors":["Li Ding","Chen Feng"],"abstract":"We propose DeepMapping, a novel registration framework using deep neural\nnetworks (DNNs) as auxiliary functions to align multiple point clouds from\nscratch to a globally consistent frame. We use DNNs to model the highly\nnon-convex mapping process that traditionally involves hand-crafted data\nassociation, sensor pose initialization, and global refinement. Our key novelty\nis that \"training\" these DNNs with properly defined unsupervised losses is\nequivalent to solving the underlying registration problem, but less sensitive\nto good initialization than ICP. Our framework contains two DNNs: a\nlocalization network that estimates the poses for input point clouds, and a map\nnetwork that models the scene structure by estimating the occupancy status of\nglobal coordinates. This allows us to convert the registration problem to a\nbinary occupancy classification, which can be solved efficiently using\ngradient-based optimization. We further show that DeepMapping can be readily\nextended to address the problem of Lidar SLAM by imposing geometric constraints\nbetween consecutive point clouds. Experiments are conducted on both simulated\nand real datasets. Qualitative and quantitative comparisons demonstrate that\nDeepMapping often enables more robust and accurate global registration of\nmultiple point clouds than existing techniques. Our code is available at\nhttps://ai4ce.github.io/DeepMapping/.","url_abs":"http://arxiv.org/abs/1811.11397v2","url_pdf":"http://arxiv.org/pdf/1811.11397v2.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":"deepmapping-unsupervised-map-estimation-from","repo_url":"https://github.com/ai4ce/DeepMapping","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"point-cloud-registration","task_name":"Point Cloud Registration"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.11397","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}