{"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/comisr-compression-informed-video-super","title":"COMISR: Compression-Informed Video Super-Resolution","arxiv_id":"2105.01237","date":"2021-05-04","proceeding":"ICCV 2021 10","authors":["Yinxiao Li","Pengchong Jin","Feng Yang","Ce Liu","Ming-Hsuan Yang","Peyman Milanfar"],"abstract":"Most video super-resolution methods focus on restoring high-resolution video frames from low-resolution videos without taking into account compression. However, most videos on the web or mobile devices are compressed, and the compression can be severe when the bandwidth is limited. In this paper, we propose a new compression-informed video super-resolution model to restore high-resolution content without introducing artifacts caused by compression. The proposed model consists of three modules for video super-resolution: bi-directional recurrent warping, detail-preserving flow estimation, and Laplacian enhancement. All these three modules are used to deal with compression properties such as the location of the intra-frames in the input and smoothness in the output frames. For thorough performance evaluation, we conducted extensive experiments on standard datasets with a wide range of compression rates, covering many real video use cases. We showed that our method not only recovers high-resolution content on uncompressed frames from the widely-used benchmark datasets, but also achieves state-of-the-art performance in super-resolving compressed videos based on numerous quantitative metrics. We also evaluated the proposed method by simulating streaming from YouTube to demonstrate its effectiveness and robustness. The source codes and trained models are available at https://github.com/google-research/google-research/tree/master/comisr.","url_abs":"https://arxiv.org/abs/2105.01237v2","url_pdf":"https://arxiv.org/pdf/2105.01237v2.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":"comisr-compression-informed-video-super","repo_url":"https://github.com/google-research/google-research","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null},{"paper_slug":"comisr-compression-informed-video-super","repo_url":"https://github.com/google-research/google-research/tree/master/comisr","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"jax","reach":null}],"tasks":[{"task_slug":"super-resolution","task_name":"Super-Resolution"},{"task_slug":"video-super-resolution","task_name":"Video Super-Resolution"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/video-super-resolution-on-msu-super-1","task":"Video Super-Resolution","dataset":"MSU Super-Resolution for Video Compression","model":"COMISR + x264","rank_in_archive_order":6,"of":85,"metrics":{"BSQ-rate over ERQA":"0.969","BSQ-rate over LPIPS":"1.118","BSQ-rate over MS-SSIM":"0.672","BSQ-rate over PSNR":"6.081","BSQ-rate over Subjective Score":"0.367","BSQ-rate over VMAF":"1.302"},"uses_additional_data":false},{"leaderboard":"/sota/video-super-resolution-on-msu-super-1","task":"Video Super-Resolution","dataset":"MSU Super-Resolution for Video Compression","model":"COMISR + vvenc","rank_in_archive_order":11,"of":85,"metrics":{"BSQ-rate over ERQA":"13.246","BSQ-rate over LPIPS":"11.026","BSQ-rate over MS-SSIM":"6.024","BSQ-rate over PSNR":"11.497","BSQ-rate over Subjective Score":"0.701","BSQ-rate over VMAF":"8.105"},"uses_additional_data":false},{"leaderboard":"/sota/video-super-resolution-on-msu-super-1","task":"Video Super-Resolution","dataset":"MSU Super-Resolution for Video Compression","model":"COMISR + x265","rank_in_archive_order":13,"of":85,"metrics":{"BSQ-rate over ERQA":"8.139","BSQ-rate over LPIPS":"12.998","BSQ-rate over MS-SSIM":"4.793","BSQ-rate over PSNR":"10.678","BSQ-rate over Subjective Score":"0.741","BSQ-rate over VMAF":"6.363"},"uses_additional_data":false},{"leaderboard":"/sota/video-super-resolution-on-msu-super-1","task":"Video Super-Resolution","dataset":"MSU Super-Resolution for Video Compression","model":"COMISR + uavs3e","rank_in_archive_order":16,"of":85,"metrics":{"BSQ-rate over ERQA":"3.427","BSQ-rate over LPIPS":"3.851","BSQ-rate over MS-SSIM":"7.711","BSQ-rate over PSNR":"5.761","BSQ-rate over Subjective Score":"1.229","BSQ-rate over VMAF":"9.47"},"uses_additional_data":false},{"leaderboard":"/sota/video-super-resolution-on-msu-super-1","task":"Video Super-Resolution","dataset":"MSU Super-Resolution for Video Compression","model":"COMISR + aomenc","rank_in_archive_order":26,"of":85,"metrics":{"BSQ-rate over ERQA":"11.177","BSQ-rate over LPIPS":"4.801","BSQ-rate over MS-SSIM":"11.303","BSQ-rate over PSNR":"15.144","BSQ-rate over Subjective Score":"1.943","BSQ-rate over VMAF":"10.67"},"uses_additional_data":false},{"leaderboard":"/sota/video-super-resolution-on-msu-vsr-benchmark","task":"Video Super-Resolution","dataset":"MSU Video Super Resolution Benchmark: Detail Restoration","model":"COMISR","rank_in_archive_order":9,"of":32,"metrics":{"1 - LPIPS":"0.879","ERQAv1.0":"0.654","FPS":"1.613","PSNR":"26.708","QRCRv1.0":"0.619","SSIM":"0.84","Subjective score":"5.637"},"uses_additional_data":false},{"leaderboard":"/sota/video-super-resolution-on-msu-video-upscalers","task":"Video Super-Resolution","dataset":"MSU Video Upscalers: Quality Enhancement","model":"COMISR","rank_in_archive_order":21,"of":48,"metrics":{"LPIPS":"0.291","PSNR":"30.97","SSIM":"0.871"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2105.01237","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2105.01237"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/google-research/google-research","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/google-research/googleresearch","reach":{"status":"gone","observed_at":"2026-09-17","how":"tree_404+repo_404"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/google-research/google-research/tree/master/comisr","reach":null}],"summary":{"unverified":3},"by_repo_kind":{"official":{"samples":3,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"62dd1717d6103648","entry":"generator_f","repo":"google-research/google-research","repo_kind":"official","path":"comisr/lib/model.py","file_url":"https://github.com/google-research/google-research/blob/HEAD/comisr/lib/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"62dd1717d6103648"}},{"code_sha256_prefix":"e77c35357f58bd9b","entry":"generator_f_decoder","repo":"google-research/google-research","repo_kind":"official","path":"comisr/lib/model.py","file_url":"https://github.com/google-research/google-research/blob/HEAD/comisr/lib/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"e77c35357f58bd9b"}},{"code_sha256_prefix":"fa51e5712feb334b","entry":"generator_f_encoder","repo":"google-research/google-research","repo_kind":"official","path":"comisr/lib/model.py","file_url":"https://github.com/google-research/google-research/blob/HEAD/comisr/lib/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"fa51e5712feb334b"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}