{"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/context-aware-query-image-representation-for","title":"Context Aware Query Image Representation for Particular Object Retrieval","arxiv_id":"1703.01226","date":"2017-03-03","proceeding":null,"authors":["Zakaria Laskar","Juho Kannala"],"abstract":"The current models of image representation based on Convolutional Neural\nNetworks (CNN) have shown tremendous performance in image retrieval. Such\nmodels are inspired by the information flow along the visual pathway in the\nhuman visual cortex. We propose that in the field of particular object\nretrieval, the process of extracting CNN representations from query images with\na given region of interest (ROI) can also be modelled by taking inspiration\nfrom human vision. Particularly, we show that by making the CNN pay attention\non the ROI while extracting query image representation leads to significant\nimprovement over the baseline methods on challenging Oxford5k and Paris6k\ndatasets. Furthermore, we propose an extension to a recently introduced\nencoding method for CNN representations, regional maximum activations of\nconvolutions (R-MAC). The proposed extension weights the regional\nrepresentations using a novel saliency measure prior to aggregation. This leads\nto further improvement in retrieval accuracy.","url_abs":"http://arxiv.org/abs/1703.01226v1","url_pdf":"http://arxiv.org/pdf/1703.01226v1.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":"context-aware-query-image-representation-for","repo_url":"https://github.com/AaltoVision/Object-Retrieval","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"caffe2","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":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}