{"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/self-attention-grid-for-person-re","title":"Self Attention Grid for Person Re-Identification","arxiv_id":"1809.08556","date":"2018-09-23","proceeding":null,"authors":["Jean-Paul Ainam","Ke Qin","Guisong Liu"],"abstract":"In this paper, we present an attention mechanism scheme to improve person\nre-identification task. Inspired by biology, we propose Self Attention Grid\n(SAG) to discover the most informative parts from a high-resolution image using\nits internal representation. In particular, given an input image, the proposed\nmodel is fed with two copies of the same image and consists of two branches.\nThe upper branch processes the high-resolution image and learns high\ndimensional feature representation while the lower branch processes the\nlow-resolution image and learn a filtering attention grid. We apply a max\nfilter operation to non-overlapping sub-regions on the high feature\nrepresentation before element-wise multiplied with the output of the second\nbranch. The feature maps of the second branch are subsequently weighted to\nreflect the importance of each patch of the grid using a softmax operation. Our\nattention module helps the network learn the most discriminative visual\nfeatures of multiple image regions and is specifically optimized to attend\nfeature representation at different levels. Extensive experiments on three\nlarge-scale datasets show that our self-attention mechanism significantly\nimproves the baseline model and outperforms various state-of-art models by a\nlarge margin.","url_abs":"http://arxiv.org/abs/1809.08556v1","url_pdf":"http://arxiv.org/pdf/1809.08556v1.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":"self-attention-grid-for-person-re","repo_url":"https://github.com/jpainam/self_attention_grid","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"person-re-identification","task_name":"Person Re-Identification"}],"methods":[{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}