{"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-colorization","title":"Deep Colorization","arxiv_id":"1605.00075","date":"2016-04-30","proceeding":"ICCV 2015 12","authors":["Zezhou Cheng","Qingxiong Yang","Bin Sheng"],"abstract":"This paper investigates into the colorization problem which converts a\ngrayscale image to a colorful version. This is a very difficult problem and\nnormally requires manual adjustment to achieve artifact-free quality. For\ninstance, it normally requires human-labelled color scribbles on the grayscale\ntarget image or a careful selection of colorful reference images (e.g.,\ncapturing the same scene in the grayscale target image). Unlike the previous\nmethods, this paper aims at a high-quality fully-automatic colorization method.\nWith the assumption of a perfect patch matching technique, the use of an\nextremely large-scale reference database (that contains sufficient color\nimages) is the most reliable solution to the colorization problem. However,\npatch matching noise will increase with respect to the size of the reference\ndatabase in practice. Inspired by the recent success in deep learning\ntechniques which provide amazing modeling of large-scale data, this paper\nre-formulates the colorization problem so that deep learning techniques can be\ndirectly employed. To ensure artifact-free quality, a joint bilateral filtering\nbased post-processing step is proposed. We further develop an adaptive image\nclustering technique to incorporate the global image information. Numerous\nexperiments demonstrate that our method outperforms the state-of-art algorithms\nboth in terms of quality and speed.","url_abs":"http://arxiv.org/abs/1605.00075v1","url_pdf":"http://arxiv.org/pdf/1605.00075v1.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-colorization","repo_url":"https://github.com/djflstkddk/Auto-Colorization","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"colorization","task_name":"Colorization"},{"task_slug":"image-clustering","task_name":"Image Clustering"},{"task_slug":"patch-matching","task_name":"Patch Matching"}],"methods":[{"method_slug":"colorization","method_name":"Colorization"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1605.00075","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}