{"links":{},"id":242243,"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00242243","sets":["1164:4619:11919:11920"]},"path":["11920"],"owner":"44499","recid":"242243","title":["Regularizing Image Encoders to Generate Bird's-Eye View Representations for Autonomous Driving Tasks"],"pubdate":{"attribute_name":"公開日","attribute_value":"2025-01-14"},"_buckets":{"deposit":"56dd9290-58c9-4065-97ab-dfde5eb9fcc5"},"_deposit":{"id":"242243","pid":{"type":"depid","value":"242243","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"Regularizing Image Encoders to Generate Bird's-Eye View Representations for Autonomous Driving Tasks","author_link":["668613","668619","668616","668620","668615","668617","668614","668618"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"Regularizing Image Encoders to Generate Bird's-Eye View Representations for Autonomous Driving Tasks"},{"subitem_title":"Regularizing Image Encoders to Generate Bird's-Eye View Representations for Autonomous Driving Tasks","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2025-01-14","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"Institute of Science Tokyo"},{"subitem_text_value":"Institute of Science Tokyo／National Institute of Informatics"},{"subitem_text_value":"Denso IT Laboratory"},{"subitem_text_value":"Institute of Science Tokyo／Denso IT Laboratory"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Institute of Science Tokyo","subitem_text_language":"en"},{"subitem_text_value":"Institute of Science Tokyo / National Institute of Informatics","subitem_text_language":"en"},{"subitem_text_value":"Denso IT Laboratory","subitem_text_language":"en"},{"subitem_text_value":"Institute of Science Tokyo / Denso IT Laboratory","subitem_text_language":"en"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"eng"}]},"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/242243/files/IPSJ-CVIM25240017.pdf","label":"IPSJ-CVIM25240017.pdf"},"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-CVIM25240017.pdf","filesize":[{"value":"1.1 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"0","billingrole":"20"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_login","version_id":"bfc3e0f6-8f58-4265-9547-5563cbdab413","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2024 by the Institute of Electronics, Information and Communication Engineers This 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":"Qiaoyi, Deng"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Satoshi, Ikehata"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Yusuke, Sekikawa"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Ikuro, Sato"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Qiaoyi, Deng","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Satoshi, Ikehata","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Yusuke, Sekikawa","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Ikuro, Sato","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_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"2188-8701","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"Bird's-Eye View (BEV) representations are critical for providing a unified spatial scene understanding to autonomous driving tasks. However, existing methods often struggle with a lack of transformation equivariance. This results in artifacts on BEV feature maps that degrade the performance of downstream tasks. To address this issue, we propose a regularization approach to enhance transformation equivariance through ego-vehicle and dynamic object motion transformations by aligning BEV features in the BEV coordinate system across consecutive frames and introduces a consistency loss to penalize feature misalignment. Experiments on the nuScenes dataset demonstrate that the proposed approach effectively reduces artifacts, stabilizes BEV representations, and improves the reliability of downstream tasks.","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Bird's-Eye View (BEV) representations are critical for providing a unified spatial scene understanding to autonomous driving tasks. However, existing methods often struggle with a lack of transformation equivariance. This results in artifacts on BEV feature maps that degrade the performance of downstream tasks. To address this issue, we propose a regularization approach to enhance transformation equivariance through ego-vehicle and dynamic object motion transformations by aligning BEV features in the BEV coordinate system across consecutive frames and introduces a consistency loss to penalize feature misalignment. Experiments on the nuScenes dataset demonstrate that the proposed approach effectively reduces artifacts, stabilizes BEV representations, and improves the reliability of downstream tasks.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"4","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":"2025-01-14","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"17","bibliographicVolumeNumber":"2025-CVIM-240"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"created":"2025-01-19T01:47:20.090635+00:00","updated":"2025-01-19T07:23:16.759205+00:00"}