{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00041310","sets":["1164:3616:3642:3644"]},"path":["3644"],"owner":"1","recid":"41310","title":["H.264符号化処理における動き予測の高速化に関する一検討"],"pubdate":{"attribute_name":"公開日","attribute_value":"2003-10-03"},"_buckets":{"deposit":"0396a5f9-991c-44a4-830f-5f1b50be75c7"},"_deposit":{"id":"41310","pid":{"type":"depid","value":"41310","revision_id":0},"owners":[1],"status":"published","created_by":1},"item_title":"H.264符号化処理における動き予測の高速化に関する一検討","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"H.264符号化処理における動き予測の高速化に関する一検討"},{"subitem_title":"Fast Motion Estimation Algorithm for H.264 Encoder","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2003-10-03","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"株式会社KDDI研究所"},{"subitem_text_value":"株式会社KDDI研究所"},{"subitem_text_value":"株式会社KDDI研究所"},{"subitem_text_value":"株式会社KDDI研究所"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"KDDI R&D Laboratories Inc.","subitem_text_language":"en"},{"subitem_text_value":"KDDI R&D Laboratories Inc.","subitem_text_language":"en"},{"subitem_text_value":"KDDI R&D Laboratories Inc.","subitem_text_language":"en"},{"subitem_text_value":"KDDI R&D Laboratories Inc.","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/41310/files/IPSJ-AVM03042004.pdf"},"date":[{"dateType":"Available","dateValue":"2005-10-03"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-AVM03042004.pdf","filesize":[{"value":"301.5 kB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"660","billingrole":"5"},{"tax":["include_tax"],"price":"330","billingrole":"6"},{"tax":["include_tax"],"price":"0","billingrole":"27"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"a5fff127-d541-4f70-8b74-4987df1d1e73","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2003 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"清水, 智行"},{"creatorName":"米山, 暁夫"},{"creatorName":"柳原, 広昌"},{"creatorName":"中島, 康之"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Tomoyuki, Shimizu","creatorNameLang":"en"},{"creatorName":"Akio, Yoneyama","creatorNameLang":"en"},{"creatorName":"Hiromasa, Yanagihara","creatorNameLang":"en"},{"creatorName":"Yasuyuki, Nakajima","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10438399","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":"H.264はMPEG-4やH.263などの従来の動画像圧縮符号化方式に比べて最大で2倍の符号化効率を目標とした符号化方式である。しかし、従来方式に比べて符号化処理の複雑さが非常に大きいため、実用化のためには符号化処理を効率よく行なうための手法が必要である。本稿では、H.264の複数ブロックサイズによる動き補償予測処理において、ブロックサイズの選択および動きベクトル探索処理を高速に行なうための手法について検討する。具体的には、最初に小さなブロックサイズでの動き探索を行ない、隣接ブロックの動きベクトルの類似性から、より大きなブロックサイズを適応的に選択し、かつ探索範囲を絞り込むことによって、H.264参照モデルJMと比較して0.1-0.4dBのPSNR低下で予測誤差計算回数を6-7%に削減することができた。","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"H.264 is a new video coding standard, which has coding efficiency about twice as much as existing standards such as MPEG-4, H.263, etc. Because it has larger computation complexity than existing standards, faster coding algorithms are desired for practical use. We propose a fast algorithm for multiple block size motion estimation. In this algorithm, motion vectors in smaller-sized block are searched at first, and a more suitable block size is chosen and search range is limited according to similarity of the motion vectors. As a result, search steps are reduced to about 6-7% of the H.264 Reference Model (JM), while loss of PSNR is at most 0.1-0.4 dB.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"28","bibliographic_titles":[{"bibliographic_title":"情報処理学会研究報告オーディオビジュアル複合情報処理（AVM）"}],"bibliographicPageStart":"23","bibliographicIssueDates":{"bibliographicIssueDate":"2003-10-03","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"99(2003-AVM-042)","bibliographicVolumeNumber":"2003"}]},"relation_version_is_last":true,"weko_creator_id":"1"},"id":41310,"updated":"2025-01-22T12:00:05.644268+00:00","links":{},"created":"2025-01-18T23:08:12.893080+00:00"}