{"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/cross-dimensional-weighting-for-aggregated","title":"Cross-dimensional Weighting for Aggregated Deep Convolutional Features","arxiv_id":"1512.04065","date":"2015-12-13","proceeding":null,"authors":["Yannis Kalantidis","Clayton Mellina","Simon Osindero"],"abstract":"We propose a simple and straightforward way of creating powerful image\nrepresentations via cross-dimensional weighting and aggregation of deep\nconvolutional neural network layer outputs. We first present a generalized\nframework that encompasses a broad family of approaches and includes\ncross-dimensional pooling and weighting steps. We then propose specific\nnon-parametric schemes for both spatial- and channel-wise weighting that boost\nthe effect of highly active spatial responses and at the same time regulate\nburstiness effects. We experiment on different public datasets for image search\nand show that our approach outperforms the current state-of-the-art for\napproaches based on pre-trained networks. We also provide an easy-to-use, open\nsource implementation that reproduces our results.","url_abs":"http://arxiv.org/abs/1512.04065v2","url_pdf":"http://arxiv.org/pdf/1512.04065v2.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":"cross-dimensional-weighting-for-aggregated","repo_url":"https://github.com/yahoo/crow","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"caffe2","reach":null}],"tasks":[{"task_slug":"image-retrieval","task_name":"Image Retrieval"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-retrieval-on-roxford-hard","task":"Image Retrieval","dataset":"ROxford (Hard)","model":"R – [O] –CroW","rank_in_archive_order":21,"of":23,"metrics":{"mAP":"13.3 "},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-roxford-medium","task":"Image Retrieval","dataset":"ROxford (Medium)","model":"R – [O] –CroW","rank_in_archive_order":20,"of":23,"metrics":{"mAP":"42.4"},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-rparis-hard","task":"Image Retrieval","dataset":"RParis (Hard)","model":"R – [O] –CroW","rank_in_archive_order":14,"of":23,"metrics":{"mAP":"47.2"},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-rparis-medium","task":"Image Retrieval","dataset":"RParis (Medium)","model":"R – [O] –CroW","rank_in_archive_order":14,"of":23,"metrics":{"mAP":"70.4"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1512.04065","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}