{"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/deep-depth-completion-of-a-single-rgb-d-image","title":"Deep Depth Completion of a Single RGB-D Image","arxiv_id":"1803.09326","date":"2018-03-25","proceeding":"CVPR 2018 6","authors":["Yinda Zhang","Thomas Funkhouser"],"abstract":"The goal of our work is to complete the depth channel of an RGB-D image.\nCommodity-grade depth cameras often fail to sense depth for shiny, bright,\ntransparent, and distant surfaces. To address this problem, we train a deep\nnetwork that takes an RGB image as input and predicts dense surface normals and\nocclusion boundaries. Those predictions are then combined with raw depth\nobservations provided by the RGB-D camera to solve for depths for all pixels,\nincluding those missing in the original observation. This method was chosen\nover others (e.g., inpainting depths directly) as the result of extensive\nexperiments with a new depth completion benchmark dataset, where holes are\nfilled in training data through the rendering of surface reconstructions\ncreated from multiview RGB-D scans. Experiments with different network inputs,\ndepth representations, loss functions, optimization methods, inpainting\nmethods, and deep depth estimation networks show that our proposed approach\nprovides better depth completions than these alternatives.","url_abs":"http://arxiv.org/abs/1803.09326v2","url_pdf":"http://arxiv.org/pdf/1803.09326v2.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":"deep-depth-completion-of-a-single-rgb-d-image","repo_url":"https://github.com/yindaz/DeepCompletionRelease","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"depth-completion","task_name":"Depth Completion"},{"task_slug":"depth-estimation","task_name":"Depth Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1803.09326","atlas_url":"https://app.syntology.ai/?focus=1803.09326","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}