{"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/hierarchical-deep-recurrent-architecture-for","title":"Hierarchical Deep Recurrent Architecture for Video Understanding","arxiv_id":"1707.03296","date":"2017-07-11","proceeding":null,"authors":["Luming Tang","Boyang Deng","Haiyu Zhao","Shuai Yi"],"abstract":"This paper introduces the system we developed for the Youtube-8M Video\nUnderstanding Challenge, in which a large-scale benchmark dataset was used for\nmulti-label video classification. The proposed framework contains hierarchical\ndeep architecture, including the frame-level sequence modeling part and the\nvideo-level classification part. In the frame-level sequence modelling part, we\nexplore a set of methods including Pooling-LSTM (PLSTM), Hierarchical-LSTM\n(HLSTM), Random-LSTM (RLSTM) in order to address the problem of large amount of\nframes in a video. We also introduce two attention pooling methods, single\nattention pooling (ATT) and multiply attention pooling (Multi-ATT) so that we\ncan pay more attention to the informative frames in a video and ignore the\nuseless frames. In the video-level classification part, two methods are\nproposed to increase the classification performance, i.e.\nHierarchical-Mixture-of-Experts (HMoE) and Classifier Chains (CC). Our final\nsubmission is an ensemble consisting of 18 sub-models. In terms of the official\nevaluation metric Global Average Precision (GAP) at 20, our best submission\nachieves 0.84346 on the public 50% of test dataset and 0.84333 on the private\n50% of test data.","url_abs":"http://arxiv.org/abs/1707.03296v1","url_pdf":"http://arxiv.org/pdf/1707.03296v1.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":"hierarchical-deep-recurrent-architecture-for","repo_url":"https://github.com/Tsingularity/youtube-8m","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"mixture-of-experts","task_name":"Mixture-of-Experts"},{"task_slug":"video-classification","task_name":"Video Classification"},{"task_slug":"video-understanding","task_name":"Video Understanding"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}