{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00083583","sets":["1164:4619:6650:6850"]},"path":["6850"],"owner":"11","recid":"83583","title":["大規模画像データセットを用いたマルチクラス物体検出器の同時学習 物体毎に特化した負例クラスの導入"],"pubdate":{"attribute_name":"公開日","attribute_value":"2012-08-26"},"_buckets":{"deposit":"4ade1217-5588-440c-97f9-497b709e669e"},"_deposit":{"id":"83583","pid":{"type":"depid","value":"83583","revision_id":0},"owners":[11],"status":"published","created_by":11},"item_title":"大規模画像データセットを用いたマルチクラス物体検出器の同時学習 物体毎に特化した負例クラスの導入","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"大規模画像データセットを用いたマルチクラス物体検出器の同時学習 物体毎に特化した負例クラスの導入"},{"subitem_title":"Simultaneous training of multi-class object detectors via large scale image dataset introduction of target specific negative classes","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2012-08-26","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"東京大学大学院情報理工学系研究科"},{"subitem_text_value":"東京大学大学院情報理工学系研究科"},{"subitem_text_value":"東京大学大学院情報理工学系研究科"},{"subitem_text_value":"東京大学大学院情報理工学系研究科"},{"subitem_text_value":"東京大学大学院情報理工学系研究科"},{"subitem_text_value":"東京大学大学院情報理工学系研究科/JSTさきがけ"},{"subitem_text_value":"東京大学大学院情報理工学系研究科"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Graduate School of Information Science and Technology, The University of Tokyo","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Information Science and Technology, The University of Tokyo","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Information Science and Technology, The University of Tokyo","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Information Science and Technology, The University of Tokyo","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Information Science and Technology, The University of Tokyo","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Information Science and Technology, The University of Tokyo / JST PRESTO","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Information Science and Technology, The University of Tokyo","subitem_text_language":"en"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_publisher":{"attribute_name":"出版者","attribute_value_mlt":[{"subitem_publisher":"情報処理学会","subitem_publisher_language":"ja"}]},"publish_status":"0","weko_shared_id":-1,"item_file_price":{"attribute_name":"Billing file","attribute_type":"file","attribute_value_mlt":[{"url":{"url":"https://ipsj.ixsq.nii.ac.jp/record/83583/files/IPSJ-CVIM12183017.pdf","label":"IPSJ-CVIM12183017"},"date":[{"dateType":"Available","dateValue":"2100-01-01"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-CVIM12183017.pdf","filesize":[{"value":"1.3 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"0","billingrole":"20"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"ef17043a-4951-47db-b022-c15ffc493041","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2012 by the Institute of Electronics, Information and Communication Engineers\nThis SIG report is only available to those in membership of the SIG."}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"金崎, 朝子"},{"creatorName":"稲葉, 翔"},{"creatorName":"牛久, 祥孝"},{"creatorName":"山下, 裕也"},{"creatorName":"村岡, 宏是"},{"creatorName":"原田, 達也"},{"creatorName":"國吉, 康夫"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Asako, Kanezaki","creatorNameLang":"en"},{"creatorName":"Sho, Inaba","creatorNameLang":"en"},{"creatorName":"Yoshitaka, Ushiku","creatorNameLang":"en"},{"creatorName":"Yuya, Yamashita","creatorNameLang":"en"},{"creatorName":"Hiroshi, Muraoka","creatorNameLang":"en"},{"creatorName":"Tatsuya, Harada","creatorNameLang":"en"},{"creatorName":"Yasuo, Kuniyoshi","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AA11131797","subitem_source_identifier_type":"NCID"}]},"item_4_textarea_12":{"attribute_name":"Notice","attribute_value_mlt":[{"subitem_textarea_value":"SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc."}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_18gh","resourcetype":"technical report"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"大規模データを用いて，マルチクラスの物体検出器を同時に最適化する効率的な手法を提案する．従来の物体検出器の学習は，各検出対象物体をpositiveサンプル，それ以外の物体をnegativeサンプルとして識別境界を決定する，”one-vs-aIl”のアプローチをとるものが主流であった．しかしながら，この方法では各クラスを独立に学習するため，異なるクラス間のスコアのバランスを調整できない．提案手法は，マルチクラスの識別手法を応用してマルチクラスの物体検出器を同時に学習することで，クラス間のバランスを最適化する．このとき，学習対象物体クラス間の差違だけでなく，その他の大量の背景画像と各クラスとの差違を考慮することで，未知物体の誤検出を抑える．実験では，大規模一般物体認識コンペテイションILSVRC2011で用いられた大量データセットのサブセットによる評価を行い，提案手法の有効性を示した．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"We propose an efficient method to train multiple object detectors simultaneously using a large-scale image dataset.The one-vs-all approach that optimizes the boundary between positive samples from a target class and negative samples from the others has been the most standard approach for object detection. However, because this approach trains each object detector independently, the likelihoods are not balanced between object classes. The proposed method combines ideas derived from both detection and classification in order to balance the scores across all object classes. We optimized the boundary between target classes and their hard-negative samples, just as in detection, while simultaneously balancing the detector likelihoods across object classes, as done in multi-class classification. We evaluated the performances on multi-class object detection using a subset of the ImageNet Large Scale Visual Recognition Challenge (ILSVRC) 2011 dataset and showed our method outperformed a de facto standard method.\n","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"8","bibliographic_titles":[{"bibliographic_title":"研究報告コンピュータビジョンとイメージメディア（CVIM）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2012-08-26","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"17","bibliographicVolumeNumber":"2012-CVIM-183"}]},"relation_version_is_last":true,"weko_creator_id":"11"},"id":83583,"updated":"2025-01-21T14:48:29.447222+00:00","links":{},"created":"2025-01-18T23:37:02.265583+00:00"}