{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,31]],"date-time":"2025-10-31T08:02:08Z","timestamp":1761897728347},"reference-count":29,"publisher":"IEEE","license":[{"start":{"date-parts":[[2022,7,18]],"date-time":"2022-07-18T00:00:00Z","timestamp":1658102400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2022,7,18]],"date-time":"2022-07-18T00:00:00Z","timestamp":1658102400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,7,18]]},"DOI":"10.1109\/icme52920.2022.9859743","type":"proceedings-article","created":{"date-parts":[[2022,8,26]],"date-time":"2022-08-26T19:45:18Z","timestamp":1661543118000},"source":"Crossref","is-referenced-by-count":3,"title":["Visual-Aware Attention Dual-Stream Decoder for Video Captioning"],"prefix":"10.1109","author":[{"given":"Zhixin","family":"Sun","sequence":"first","affiliation":[{"name":"School of Computer Science and Artificial Intelligence, Wuhan University of Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shuqin","family":"Chen","sequence":"additional","affiliation":[{"name":"School of Computer Science and Artificial Intelligence, Wuhan University of Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Luo","family":"Zhong","sequence":"additional","affiliation":[{"name":"School of Computer Science and Artificial Intelligence, Wuhan University of Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1007\/s11063-020-10352-2"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01088"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33018191"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/WACV45572.2020.9093291"},{"key":"ref14","article-title":"At-tention on attention for image captioning","author":"huang","year":"2019","journal-title":"ICCV"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00852"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1145\/3343031.3351072"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00856"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2020\/104"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D17-1103"},{"key":"ref28","doi-asserted-by":"crossref","DOI":"10.1609\/aaai.v35i3.16353","article-title":"Semantic grouping network for video captioning","author":"ryu","year":"2021","journal-title":"AAAI"},{"key":"ref4","article-title":"Bridging the gap be-tween training and inference for neural machine translation (extended abstract)","author":"zhang","year":"2020","journal-title":"IJCAI"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01311"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2022.3141886"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2020\/88"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00901"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33018271"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00854"},{"key":"ref7","article-title":"Sibnet: Sibling con-volutional encoder for video captioning","author":"liu","year":"2018","journal-title":"ACM MM"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2021.3102504"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1145\/3343031.3350932"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00647"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00273"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1145\/3343031.3351060"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33018167"},{"key":"ref24","article-title":"Hierarchical lstms with adaptive attention for visual captioning","author":"gao","year":"2020","journal-title":"TPAMI"},{"key":"ref23","article-title":"Spatio-temporal dynamics and seman-tic attribute enriched visual encoding for video captioning","author":"aafaq","year":"2019","journal-title":"CVPR"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01088"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1145\/3394171.3416291"}],"event":{"name":"2022 IEEE International Conference on Multimedia and Expo (ICME)","location":"Taipei, Taiwan","start":{"date-parts":[[2022,7,18]]},"end":{"date-parts":[[2022,7,22]]}},"container-title":["2022 IEEE International Conference on Multimedia and Expo (ICME)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9859562\/9858923\/09859743.pdf?arnumber=9859743","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,2,16]],"date-time":"2023-02-16T03:20:58Z","timestamp":1676517658000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9859743\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,7,18]]},"references-count":29,"URL":"https:\/\/doi.org\/10.1109\/icme52920.2022.9859743","relation":{},"subject":[],"published":{"date-parts":[[2022,7,18]]}}}