{"id":223072,"updated":"2025-01-19T13:31:41.500562+00:00","links":{},"created":"2025-01-19T01:22:56.736073+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00223072","sets":["6164:6165:6462:11124"]},"path":["11124"],"owner":"44499","recid":"223072","title":["LightInceptionNet : 高精度で軽量なDeepfake検出技術の開発"],"pubdate":{"attribute_name":"公開日","attribute_value":"2022-10-17"},"_buckets":{"deposit":"3f77b484-33f4-4636-bba9-88eb7db7820e"},"_deposit":{"id":"223072","pid":{"type":"depid","value":"223072","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"LightInceptionNet : 高精度で軽量なDeepfake検出技術の開発","author_link":["586725","586729","586728","586724","586726","586727"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"LightInceptionNet : 高精度で軽量なDeepfake検出技術の開発"},{"subitem_title":"LightInceptionNet : Development of Accurate and Lightweight Deepfake Detection Scheme","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"Deepfake，検出，画像認識，深層学習","subitem_subject_scheme":"Other"}]},"item_type_id":"18","publish_date":"2022-10-17","item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_18_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"早稲田大学"},{"subitem_text_value":"早稲田大学"},{"subitem_text_value":"早稲田大学／NICT／理研AIP"}]},"item_18_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Waseda University","subitem_text_language":"en"},{"subitem_text_value":"Waseda University","subitem_text_language":"en"},{"subitem_text_value":"Waseda University / NICT / RIKEN AIP","subitem_text_language":"en"}]},"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/223072/files/IPSJ-CSS2022017.pdf","label":"IPSJ-CSS2022017.pdf"},"date":[{"dateType":"Available","dateValue":"2024-10-17"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-CSS2022017.pdf","filesize":[{"value":"3.2 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"660","billingrole":"5"},{"tax":["include_tax"],"price":"330","billingrole":"6"},{"tax":["include_tax"],"price":"0","billingrole":"30"},{"tax":["include_tax"],"price":"0","billingrole":"46"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"5378492e-0bdc-46ed-94be-826a7ac63b13","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2022 by the Information Processing Society of Japan"}]},"item_18_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"利川, 悠斗"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"飯島, 涼"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"森, 達哉"}],"nameIdentifiers":[{}]}]},"item_18_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Yuto, Toshikawa","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Ryo, Iijima","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Tatsuya, Mori","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_5794","resourcetype":"conference paper"}]},"item_18_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"本研究は，Deepfake を高精度に検出可能かつ軽量な検出モデルの開発を目的とする．具体的には，画像認識用アーキテクチャである InceptionNet を軽量化した LightInceptionNet を提案し，LightInceptionNet を用いた Deepfake 検知モデルを作成する．LightInceptionNet の中心となるアイディアは，パラメータ数の削減が可能な畳み込み層 (SeparableConv2D) を導入することであり，精度を保ちつつパラメータ数を削減できる．Celeb-DF データセットを用いた性能評価の結果，テスト精度 (Accuracy) が 88.46%，テスト AUC Score が 93.82% を達成し，既存の SoTA 実装と比較して高精度に Deepfake を検出可能である．また，顔部分抽出の前処理の導入により，学習時間を 22.46% 削減した．本検知手法により，Deepfake を悪用した社会問題を防止する効果が期待できる．","subitem_description_type":"Other"}]},"item_18_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"The goal of this work is to develop a lightweight detection model that can detect deepfakes with high accuracy. We propose LightInceptionNet --- a lightweight version of InceptionNet, which is an architecture for image recognition. We build a deepfake detection model using the LightInceptionNet. The core idea of LightInceptionNet is to introduce a convolutional layer (SeparableConv2D) that can reduce the number of parameters  while maintaining accuracy. Our performance evaluation using the Celeb-DF dataset demonstrates that the test accuracy is 88.46% and the test AUC score is 93.82%. In addition, the introduction of preprocessing for facial extraction reduced the learning time by 22.46%. This detection method can be expected to prevent social problems that abuse Deepfake.\n","subitem_description_type":"Other"}]},"item_18_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"112","bibliographic_titles":[{"bibliographic_title":"コンピュータセキュリティシンポジウム2022論文集"}],"bibliographicPageStart":"105","bibliographicIssueDates":{"bibliographicIssueDate":"2022-10-17","bibliographicIssueDateType":"Issued"}}]},"relation_version_is_last":true,"weko_creator_id":"44499"}}