{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00211198","sets":["1164:4619:10416:10591"]},"path":["10591"],"owner":"44499","recid":"211198","title":["自己教師あり学習を用いたシングルモーダルデータによる地面材質のクラスタリング"],"pubdate":{"attribute_name":"公開日","attribute_value":"2021-05-13"},"_buckets":{"deposit":"04e51555-9b58-4d60-ab09-cff143f5ad48"},"_deposit":{"id":"211198","pid":{"type":"depid","value":"211198","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"自己教師あり学習を用いたシングルモーダルデータによる地面材質のクラスタリング","author_link":["536175","536176","536173","536174"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"自己教師あり学習を用いたシングルモーダルデータによる地面材質のクラスタリング"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"卒論スポットライトセッション","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2021-05-13","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"慶應義塾大学"},{"subitem_text_value":"慶應義塾大学"},{"subitem_text_value":"慶應義塾大学"},{"subitem_text_value":"慶應義塾大学"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Department of Information and Computer Science, Keio University","subitem_text_language":"en"},{"subitem_text_value":"Department of Information and Computer Science, Keio University","subitem_text_language":"en"},{"subitem_text_value":"Department of Information and Computer Science, Keio University","subitem_text_language":"en"},{"subitem_text_value":"Department of Information and Computer Science, Keio University","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/211198/files/IPSJ-CVIM21226041.pdf","label":"IPSJ-CVIM21226041.pdf"},"date":[{"dateType":"Available","dateValue":"2023-05-13"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-CVIM21226041.pdf","filesize":[{"value":"1.6 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":"20"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"e1878b2f-7e6e-4cbd-add3-bf33ab81682a","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2021 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"石川, 玲奈"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"八馬, 遼"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"黒部, 聡亮"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"斎藤, 英雄"}],"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":"地形を正確に把握するためには，さまざまなセンサーから得られるマルチモーダルデータから有益な特徴量を抽出することが重要である．RGB カメラやデプスセンサ，振動センサー，マイクロフォンなどがそのセンサーの例である．特にロボティクス分野では，シングルモーダル，あるいはマルチモーダルな手法を導入して特徴量の抽出に成功した論文は複数が存在する．しかし実場面において，マイクが機能しないような混雑時，あるいは RGB カメラが機能しないような暗闇の中というような extreme conditions と呼ばれる場面にロボットは直面することがあり，既存研究においてはこのような場面が考慮されていないことが現状である．そこで本論文では，マルチモーダル変分オートエンコーダと混合ガウスモデルを用いて，画像データと音データを対象とした，地面の材質のクラスタリングのための新しいフレームワークを提案する．この手法はテスト時に画像または音声のいずれかのモダリティが欠落していてもクラスタリングを可能にする手法である．さらに我々は提案手法の有効性を示すため，従来のマルチモーダルな材質のクラスタリング手法を用いてクラスタリング精度を評価し，いくつかの ablation study を行った．","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":"2021-05-13","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"41","bibliographicVolumeNumber":"2021-CVIM-226"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":211198,"updated":"2025-01-19T17:52:52.391643+00:00","links":{},"created":"2025-01-19T01:12:23.098050+00:00"}