{"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/mganet-a-robust-model-for-quality-enhancement","title":"MGANet: A Robust Model for Quality Enhancement of Compressed Video","arxiv_id":"1811.09150","date":"2018-11-22","proceeding":null,"authors":["Xiandong Meng","Xuan Deng","Shuyuan Zhu","Shuaicheng Liu","Chuan Wang","Chen Chen","Bing Zeng"],"abstract":"In video compression, most of the existing deep learning approaches\nconcentrate on the visual quality of a single frame, while ignoring the useful\npriors as well as the temporal information of adjacent frames. In this paper,\nwe propose a multi-frame guided attention network (MGANet) to enhance the\nquality of compressed videos. Our network is composed of a temporal encoder\nthat discovers inter-frame relations, a guided encoder-decoder subnet that\nencodes and enhances the visual patterns of target frame, and a\nmulti-supervised reconstruction component that aggregates information to\npredict details. We design a bidirectional residual convolutional LSTM unit to\nimplicitly discover frames variations over time with respect to the target\nframe. Meanwhile, the guided map is proposed to guide our network to\nconcentrate more on the block boundary. Our approach takes advantage of\nintra-frame prior information and inter-frame information to improve the\nquality of compressed video. Experimental results show the robustness and\nsuperior performance of the proposed method.Code is available at\nhttps://github.com/mengab/MGANet","url_abs":"http://arxiv.org/abs/1811.09150v4","url_pdf":"http://arxiv.org/pdf/1811.09150v4.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":"mganet-a-robust-model-for-quality-enhancement","repo_url":"https://github.com/mengab/MGANet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"mganet-a-robust-model-for-quality-enhancement","repo_url":"https://github.com/mengab/MGANet-DCC2020","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"mganet-a-robust-model-for-quality-enhancement","repo_url":"https://github.com/MindSpore-scientific/code-11/tree/main/mgan-mindspore","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"mganet-a-robust-model-for-quality-enhancement","repo_url":"https://github.com/MindSpore-scientific/code-8/tree/main/mgan-mindspore","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"video-compression","task_name":"Video Compression"}],"methods":[{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}