{"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/few-example-object-detection-with-model","title":"Few-Example Object Detection with Model Communication","arxiv_id":"1706.08249","date":"2017-06-26","proceeding":null,"authors":["Xuanyi Dong","Liang Zheng","Fan Ma","Yi Yang","Deyu Meng"],"abstract":"In this paper, we study object detection using a large pool of unlabeled\nimages and only a few labeled images per category, named \"few-example object\ndetection\". The key challenge consists in generating trustworthy training\nsamples as many as possible from the pool. Using few training examples as\nseeds, our method iterates between model training and high-confidence sample\nselection. In training, easy samples are generated first and, then the poorly\ninitialized model undergoes improvement. As the model becomes more\ndiscriminative, challenging but reliable samples are selected. After that,\nanother round of model improvement takes place. To further improve the\nprecision and recall of the generated training samples, we embed multiple\ndetection models in our framework, which has proven to outperform the single\nmodel baseline and the model ensemble method. Experiments on PASCAL VOC'07, MS\nCOCO'14, and ILSVRC'13 indicate that by using as few as three or four samples\nselected for each category, our method produces very competitive results when\ncompared to the state-of-the-art weakly-supervised approaches using a large\nnumber of image-level labels.","url_abs":"http://arxiv.org/abs/1706.08249v8","url_pdf":"http://arxiv.org/pdf/1706.08249v8.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":"few-example-object-detection-with-model","repo_url":"https://github.com/D-X-Y/DXY-Projects/tree/master/MSPLD","is_official":0,"mentioned_in_paper":0,"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":"model","task_name":"model"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/weakly-supervised-object-detection-on-coco","task":"Weakly Supervised Object Detection","dataset":"COCO (Common Objects in Context)","model":"MSLPD","rank_in_archive_order":1,"of":5,"metrics":{"MAP":"56.6"},"uses_additional_data":false},{"leaderboard":"/sota/weakly-supervised-object-detection-on","task":"Weakly Supervised Object Detection","dataset":"ImageNet","model":"MSLPD","rank_in_archive_order":3,"of":4,"metrics":{"MAP":"13.9"},"uses_additional_data":false},{"leaderboard":"/sota/weakly-supervised-object-detection-on-pascal-1","task":"Weakly Supervised Object Detection","dataset":"PASCAL VOC 2007","model":"MSLPD","rank_in_archive_order":34,"of":41,"metrics":{"MAP":"41.7"},"uses_additional_data":false},{"leaderboard":"/sota/weakly-supervised-object-detection-on-pascal","task":"Weakly Supervised Object Detection","dataset":"PASCAL VOC 2012 test","model":"MSLPD","rank_in_archive_order":30,"of":32,"metrics":{"MAP":"35.4"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1706.08249","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}