{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/lightweight-attentional-feature-fusion-for","title":"Lightweight Attentional Feature Fusion: A New Baseline for Text-to-Video Retrieval","arxiv_id":"2112.01832","date":"2021-12-03","proceeding":null,"authors":["Fan Hu","Aozhu Chen","Ziyue Wang","Fangming Zhou","Jianfeng Dong","Xirong Li"],"abstract":"In this paper we revisit feature fusion, an old-fashioned topic, in the new context of text-to-video retrieval. Different from previous research that considers feature fusion only at one end, let it be video or text, we aim for feature fusion for both ends within a unified framework. We hypothesize that optimizing the convex combination of the features is preferred to modeling their correlations by computationally heavy multi-head self attention. We propose Lightweight Attentional Feature Fusion (LAFF). LAFF performs feature fusion at both early and late stages and at both video and text ends, making it a powerful method for exploiting diverse (off-the-shelf) features. The interpretability of LAFF can be used for feature selection. Extensive experiments on five public benchmark sets (MSR-VTT, MSVD, TGIF, VATEX and TRECVID AVS 2016-2020) justify LAFF as a new baseline for text-to-video retrieval.","url_abs":"https://arxiv.org/abs/2112.01832v3","url_pdf":"https://arxiv.org/pdf/2112.01832v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"lightweight-attentional-feature-fusion-for","repo_url":"https://github.com/ruc-aimc-lab/laff","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"ad-hoc-video-search","task_name":"Ad-hoc video search"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"text-to-video-retrieval","task_name":"Text to Video Retrieval"},{"task_slug":"video-retrieval","task_name":"Video Retrieval"},{"task_slug":"feature-selection","task_name":"feature selection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/ad-hoc-video-search-on-trecvid-avs16-iacc-3","task":"Ad-hoc video search","dataset":"TRECVID-AVS16 (IACC.3)","model":"LAFF","rank_in_archive_order":1,"of":4,"metrics":{"infAP":"0.222"},"uses_additional_data":true},{"leaderboard":"/sota/ad-hoc-video-search-on-trecvid-avs17-iacc-3","task":"Ad-hoc video search","dataset":"TRECVID-AVS17 (IACC.3)","model":"LAFF","rank_in_archive_order":1,"of":4,"metrics":{"infAP":"0.290"},"uses_additional_data":true},{"leaderboard":"/sota/ad-hoc-video-search-on-trecvid-avs18-iacc-3","task":"Ad-hoc video search","dataset":"TRECVID-AVS18 (IACC.3)","model":"LAFF","rank_in_archive_order":1,"of":4,"metrics":{"infAP":"0.147"},"uses_additional_data":true},{"leaderboard":"/sota/ad-hoc-video-search-on-trecvid-avs19-v3c1","task":"Ad-hoc video search","dataset":"TRECVID-AVS19 (V3C1)","model":"LAFF","rank_in_archive_order":1,"of":2,"metrics":{"infAP":"0.192"},"uses_additional_data":true},{"leaderboard":"/sota/ad-hoc-video-search-on-trecvid-avs20-v3c1","task":"Ad-hoc video search","dataset":"TRECVID-AVS20 (V3C1)","model":"LAFF","rank_in_archive_order":1,"of":1,"metrics":{"infAP":"0.265"},"uses_additional_data":true},{"leaderboard":"/sota/video-retrieval-on-msr-vtt","task":"Video Retrieval","dataset":"MSR-VTT","model":"LAFF","rank_in_archive_order":27,"of":40,"metrics":{"text-to-video R@1":"29.1","text-to-video R@10":"65.8","text-to-video R@5":"54.9"},"uses_additional_data":false},{"leaderboard":"/sota/video-retrieval-on-msr-vtt-1ka","task":"Video Retrieval","dataset":"MSR-VTT-1kA","model":"LAFF","rank_in_archive_order":35,"of":63,"metrics":{"text-to-video R@1":"45.8","text-to-video R@10":"82","text-to-video R@5":"71.5"},"uses_additional_data":false},{"leaderboard":"/sota/video-retrieval-on-msvd","task":"Video Retrieval","dataset":"MSVD","model":"LAFF","rank_in_archive_order":20,"of":24,"metrics":{"text-to-video R@1":"45.4","text-to-video R@10":"84.6","text-to-video R@5":"76.0"},"uses_additional_data":false},{"leaderboard":"/sota/video-retrieval-on-tgif","task":"Video Retrieval","dataset":"TGIF","model":"LAFF","rank_in_archive_order":2,"of":2,"metrics":{"text-to-video R@1":"24.5","text-to-video R@10":"54.5","text-to-video R@5":"45.0"},"uses_additional_data":false},{"leaderboard":"/sota/video-retrieval-on-vatex","task":"Video Retrieval","dataset":"VATEX","model":"LAFF","rank_in_archive_order":11,"of":13,"metrics":{"text-to-video R@1":"59.1","text-to-video R@10":"91.7","text-to-video R@50":"96.3"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2112.01832","atlas_url":"https://app.syntology.ai/?focus=2112.01832","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}