{"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/fastmask-segment-multi-scale-object","title":"FastMask: Segment Multi-scale Object Candidates in One Shot","arxiv_id":"1612.08843","date":"2016-12-28","proceeding":"CVPR 2017 7","authors":["Hexiang Hu","Shiyi Lan","Yuning Jiang","Zhimin Cao","Fei Sha"],"abstract":"Objects appear to scale differently in natural images. This fact requires\nmethods dealing with object-centric tasks (e.g. object proposal) to have robust\nperformance over variances in object scales. In the paper, we present a novel\nsegment proposal framework, namely FastMask, which takes advantage of\nhierarchical features in deep convolutional neural networks to segment\nmulti-scale objects in one shot. Innovatively, we adapt segment proposal\nnetwork into three different functional components (body, neck and head). We\nfurther propose a weight-shared residual neck module as well as a\nscale-tolerant attentional head module for efficient one-shot inference. On MS\nCOCO benchmark, the proposed FastMask outperforms all state-of-the-art segment\nproposal methods in average recall being 2~5 times faster. Moreover, with a\nslight trade-off in accuracy, FastMask can segment objects in near real time\n(~13 fps) with 800*600 resolution images, demonstrating its potential in\npractical applications. Our implementation is available on\nhttps://github.com/voidrank/FastMask.","url_abs":"http://arxiv.org/abs/1612.08843v4","url_pdf":"http://arxiv.org/pdf/1612.08843v4.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":"fastmask-segment-multi-scale-object","repo_url":"https://github.com/voidrank/FastMask","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"caffe2","reach":null},{"paper_slug":"fastmask-segment-multi-scale-object","repo_url":"https://github.com/chwilms/AttentionMask","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"fastmask-segment-multi-scale-object","repo_url":"https://github.com/chwilms/superpixelRefinement","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"object","task_name":"Object"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1612.08843","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}