{"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/crafting-gbd-net-for-object-detection","title":"Crafting GBD-Net for Object Detection","arxiv_id":"1610.02579","date":"2016-10-08","proceeding":null,"authors":["Xingyu Zeng","Wanli Ouyang","Junjie Yan","Hongsheng Li","Tong Xiao","Kun Wang","Yu Liu","Yucong Zhou","Bin Yang","Zhe Wang","Hui Zhou","Xiaogang Wang"],"abstract":"The visual cues from multiple support regions of different sizes and\nresolutions are complementary in classifying a candidate box in object\ndetection. Effective integration of local and contextual visual cues from these\nregions has become a fundamental problem in object detection.\n  In this paper, we propose a gated bi-directional CNN (GBD-Net) to pass\nmessages among features from different support regions during both feature\nlearning and feature extraction. Such message passing can be implemented\nthrough convolution between neighboring support regions in two directions and\ncan be conducted in various layers. Therefore, local and contextual visual\npatterns can validate the existence of each other by learning their nonlinear\nrelationships and their close interactions are modeled in a more complex way.\nIt is also shown that message passing is not always helpful but dependent on\nindividual samples. Gated functions are therefore needed to control message\ntransmission, whose on-or-offs are controlled by extra visual evidence from the\ninput sample. The effectiveness of GBD-Net is shown through experiments on\nthree object detection datasets, ImageNet, Pascal VOC2007 and Microsoft COCO.\nThis paper also shows the details of our approach in wining the ImageNet object\ndetection challenge of 2016, with source code provided on\n\\url{https://github.com/craftGBD/craftGBD}.","url_abs":"http://arxiv.org/abs/1610.02579v1","url_pdf":"http://arxiv.org/pdf/1610.02579v1.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":"crafting-gbd-net-for-object-detection","repo_url":"https://github.com/craftGBD/craftGBD","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}