{"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/towards-visual-feature-translation","title":"Towards Visual Feature Translation","arxiv_id":"1812.00573","date":"2018-12-03","proceeding":"CVPR 2019 6","authors":["Jie Hu","Rongrong Ji","Hong Liu","Shengchuan Zhang","Cheng Deng","Qi Tian"],"abstract":"Most existing visual search systems are deployed based upon fixed kinds of\nvisual features, which prohibits the feature reusing across different systems\nor when upgrading systems with a new type of feature. Such a setting is\nobviously inflexible and time/memory consuming, which is indeed mendable if\nvisual features can be \"translated\" across systems. In this paper, we make the\nfirst attempt towards visual feature translation to break through the barrier\nof using features across different visual search systems. To this end, we\npropose a Hybrid Auto-Encoder (HAE) to translate visual features, which learns\na mapping by minimizing the translation and reconstruction errors. Based upon\nHAE, an Undirected Affinity Measurement (UAM) is further designed to quantify\nthe affinity among different types of visual features. Extensive experiments\nhave been conducted on several public datasets with sixteen different types of\nwidely-used features in visual search systems. Quantitative results show the\nencouraging possibilities of feature translation. For the first time, the\naffinity among widely-used features like SIFT and DELF is reported.","url_abs":"http://arxiv.org/abs/1812.00573v2","url_pdf":"http://arxiv.org/pdf/1812.00573v2.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":"towards-visual-feature-translation","repo_url":"https://github.com/hujiecpp/VisualFeatureTranslation","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"translation","task_name":"Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1812.00573","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}