{"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/selective-deep-convolutional-features-for","title":"Selective Deep Convolutional Features for Image Retrieval","arxiv_id":"1707.00809","date":"2017-07-04","proceeding":null,"authors":["Tuan Hoang","Thanh-Toan Do","Dang-Khoa Le Tan","Ngai-Man Cheung"],"abstract":"Convolutional Neural Network (CNN) is a very powerful approach to extract\ndiscriminative local descriptors for effective image search. Recent work adopts\nfine-tuned strategies to further improve the discriminative power of the\ndescriptors. Taking a different approach, in this paper, we propose a novel\nframework to achieve competitive retrieval performance. Firstly, we propose\nvarious masking schemes, namely SIFT-mask, SUM-mask, and MAX-mask, to select a\nrepresentative subset of local convolutional features and remove a large number\nof redundant features. We demonstrate that this can effectively address the\nburstiness issue and improve retrieval accuracy. Secondly, we propose to employ\nrecent embedding and aggregating methods to further enhance feature\ndiscriminability. Extensive experiments demonstrate that our proposed framework\nachieves state-of-the-art retrieval accuracy.","url_abs":"http://arxiv.org/abs/1707.00809v2","url_pdf":"http://arxiv.org/pdf/1707.00809v2.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":"selective-deep-convolutional-features-for","repo_url":"https://github.com/hnanhtuan/selectiveConvFeature","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"image-retrieval","task_name":"Image Retrieval"},{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1707.00809","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}