{"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/fast-spectral-ranking-for-similarity-search","title":"Fast Spectral Ranking for Similarity Search","arxiv_id":"1703.06935","date":"2017-03-20","proceeding":"CVPR 2018 6","authors":["Ahmet Iscen","Yannis Avrithis","Giorgos Tolias","Teddy Furon","Ondrej Chum"],"abstract":"Despite the success of deep learning on representing images for particular\nobject retrieval, recent studies show that the learned representations still\nlie on manifolds in a high dimensional space. This makes the Euclidean nearest\nneighbor search biased for this task. Exploring the manifolds online remains\nexpensive even if a nearest neighbor graph has been computed offline. This work\nintroduces an explicit embedding reducing manifold search to Euclidean search\nfollowed by dot product similarity search. This is equivalent to linear graph\nfiltering of a sparse signal in the frequency domain. To speed up online\nsearch, we compute an approximate Fourier basis of the graph offline. We\nimprove the state of art on particular object retrieval datasets including the\nchallenging Instre dataset containing small objects. At a scale of 10^5 images,\nthe offline cost is only a few hours, while query time is comparable to\nstandard similarity search.","url_abs":"http://arxiv.org/abs/1703.06935v3","url_pdf":"http://arxiv.org/pdf/1703.06935v3.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":"fast-spectral-ranking-for-similarity-search","repo_url":"https://github.com/ducha-aiki/manifold-diffusion","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}