{"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/in-defense-of-classical-image-processing-fast","title":"In Defense of Classical Image Processing: Fast Depth Completion on the CPU","arxiv_id":"1802.00036","date":"2018-01-31","proceeding":null,"authors":["Jason Ku","Ali Harakeh","Steven L. Waslander"],"abstract":"With the rise of data driven deep neural networks as a realization of\nuniversal function approximators, most research on computer vision problems has\nmoved away from hand crafted classical image processing algorithms. This paper\nshows that with a well designed algorithm, we are capable of outperforming\nneural network based methods on the task of depth completion. The proposed\nalgorithm is simple and fast, runs on the CPU, and relies only on basic image\nprocessing operations to perform depth completion of sparse LIDAR depth data.\nWe evaluate our algorithm on the challenging KITTI depth completion benchmark,\nand at the time of submission, our method ranks first on the KITTI test server\namong all published methods. Furthermore, our algorithm is data independent,\nrequiring no training data to perform the task at hand. The code written in\nPython will be made publicly available at https://github.com/kujason/ip_basic.","url_abs":"http://arxiv.org/abs/1802.00036v1","url_pdf":"http://arxiv.org/pdf/1802.00036v1.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":"in-defense-of-classical-image-processing-fast","repo_url":"https://github.com/kujason/ip_basic","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"in-defense-of-classical-image-processing-fast","repo_url":"https://github.com/sabadijou/FastDepthCompletionCPU","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":null,"task_name":"CPU"},{"task_slug":"depth-completion","task_name":"Depth Completion"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1802.00036","atlas_url":"https://app.syntology.ai/?focus=1802.00036","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}