{"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/class-weighted-convolutional-features-for","title":"Class-Weighted Convolutional Features for Visual Instance Search","arxiv_id":"1707.02581","date":"2017-07-09","proceeding":null,"authors":["Albert Jimenez","Jose M. Alvarez","Xavier Giro-i-Nieto"],"abstract":"Image retrieval in realistic scenarios targets large dynamic datasets of\nunlabeled images. In these cases, training or fine-tuning a model every time\nnew images are added to the database is neither efficient nor scalable.\nConvolutional neural networks trained for image classification over large\ndatasets have been proven effective feature extractors for image retrieval. The\nmost successful approaches are based on encoding the activations of\nconvolutional layers, as they convey the image spatial information. In this\npaper, we go beyond this spatial information and propose a local-aware encoding\nof convolutional features based on semantic information predicted in the target\nimage. To this end, we obtain the most discriminative regions of an image using\nClass Activation Maps (CAMs). CAMs are based on the knowledge contained in the\nnetwork and therefore, our approach, has the additional advantage of not\nrequiring external information. In addition, we use CAMs to generate object\nproposals during an unsupervised re-ranking stage after a first fast search.\nOur experiments on two public available datasets for instance retrieval,\nOxford5k and Paris6k, demonstrate the competitiveness of our approach\noutperforming the current state-of-the-art when using off-the-shelf models\ntrained on ImageNet. The source code and model used in this paper are publicly\navailable at http://imatge-upc.github.io/retrieval-2017-cam/.","url_abs":"http://arxiv.org/abs/1707.02581v1","url_pdf":"http://arxiv.org/pdf/1707.02581v1.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":"class-weighted-convolutional-features-for","repo_url":"https://github.com/imatge-upc/retrieval-2017-cam","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"class-weighted-convolutional-features-for","repo_url":"https://github.com/zxy14120448/Summary","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"image-retrieval","task_name":"Image Retrieval"},{"task_slug":"instance-search","task_name":"Instance Search"},{"task_slug":"re-ranking","task_name":"Re-Ranking"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}