{"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/reference-based-video-super-resolution-using","title":"Reference-based Video Super-Resolution Using Multi-Camera Video Triplets","arxiv_id":"2203.14537","date":"2022-03-28","proceeding":"CVPR 2022 1","authors":["Junyong Lee","Myeonghee Lee","Sunghyun Cho","Seungyong Lee"],"abstract":"We propose the first reference-based video super-resolution (RefVSR) approach that utilizes reference videos for high-fidelity results. We focus on RefVSR in a triple-camera setting, where we aim at super-resolving a low-resolution ultra-wide video utilizing wide-angle and telephoto videos. We introduce the first RefVSR network that recurrently aligns and propagates temporal reference features fused with features extracted from low-resolution frames. To facilitate the fusion and propagation of temporal reference features, we propose a propagative temporal fusion module. For learning and evaluation of our network, we present the first RefVSR dataset consisting of triplets of ultra-wide, wide-angle, and telephoto videos concurrently taken from triple cameras of a smartphone. We also propose a two-stage training strategy fully utilizing video triplets in the proposed dataset for real-world 4x video super-resolution. We extensively evaluate our method, and the result shows the state-of-the-art performance in 4x super-resolution.","url_abs":"https://arxiv.org/abs/2203.14537v1","url_pdf":"https://arxiv.org/pdf/2203.14537v1.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":"reference-based-video-super-resolution-using","repo_url":"https://github.com/codeslake/RefVSR","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"AGPL-3.0"}}],"tasks":[{"task_slug":"reference-based-video-super-resolution","task_name":"Reference-based Video Super-Resolution"},{"task_slug":"super-resolution","task_name":"Super-Resolution"},{"task_slug":"video-super-resolution","task_name":"Video Super-Resolution"}],"methods":[],"datasets_introduced":[{"slug":"realmcvsr-dataset","name":"RealMCVSR","full_name":"Real-world Multi-Camera Video Super-Resolution"}],"methods_introduced":[],"results":[{"leaderboard":"/sota/reference-based-video-super-resolution-on","task":"Reference-based Video Super-Resolution","dataset":"RealMCVSR","model":"RefVSR-IR-ℓ1","rank_in_archive_order":1,"of":10,"metrics":{"PSNR":"34.86"},"uses_additional_data":false},{"leaderboard":"/sota/reference-based-video-super-resolution-on","task":"Reference-based Video Super-Resolution","dataset":"RealMCVSR","model":"RefVSR-ℓ1","rank_in_archive_order":2,"of":10,"metrics":{"PSNR":"34.74"},"uses_additional_data":false},{"leaderboard":"/sota/reference-based-video-super-resolution-on","task":"Reference-based Video Super-Resolution","dataset":"RealMCVSR","model":"RefVSR-small-ℓ1","rank_in_archive_order":3,"of":10,"metrics":{"PSNR":"33.88"},"uses_additional_data":false},{"leaderboard":"/sota/reference-based-video-super-resolution-on","task":"Reference-based Video Super-Resolution","dataset":"RealMCVSR","model":"IconVSR-ℓch [chan2021basicvsr]","rank_in_archive_order":4,"of":10,"metrics":{"PSNR":"33.80"},"uses_additional_data":false},{"leaderboard":"/sota/reference-based-video-super-resolution-on","task":"Reference-based Video Super-Resolution","dataset":"RealMCVSR","model":"BasicVSR-ℓch [chan2021basicvsr]","rank_in_archive_order":5,"of":10,"metrics":{"PSNR":"33.66"},"uses_additional_data":false},{"leaderboard":"/sota/reference-based-video-super-resolution-on","task":"Reference-based Video Super-Resolution","dataset":"RealMCVSR","model":"EDVR-ℓch [wang2019edvr]","rank_in_archive_order":6,"of":10,"metrics":{"PSNR":"33.47"},"uses_additional_data":false},{"leaderboard":"/sota/reference-based-video-super-resolution-on","task":"Reference-based Video Super-Resolution","dataset":"RealMCVSR","model":"EDVR-M-ℓch [wang2019edvr]","rank_in_archive_order":7,"of":10,"metrics":{"PSNR":"33.26"},"uses_additional_data":false},{"leaderboard":"/sota/reference-based-video-super-resolution-on","task":"Reference-based Video Super-Resolution","dataset":"RealMCVSR","model":"DCSR-ℓ1 [wang2021DCSR]","rank_in_archive_order":8,"of":10,"metrics":{"PSNR":"32.43"},"uses_additional_data":false},{"leaderboard":"/sota/reference-based-video-super-resolution-on","task":"Reference-based Video Super-Resolution","dataset":"RealMCVSR","model":"RCAN-ℓ1 [zhang2018rcan]","rank_in_archive_order":9,"of":10,"metrics":{"PSNR":"31.07"},"uses_additional_data":false},{"leaderboard":"/sota/reference-based-video-super-resolution-on","task":"Reference-based Video Super-Resolution","dataset":"RealMCVSR","model":"TTSR-ℓ1 [yang2020TTSR]","rank_in_archive_order":10,"of":10,"metrics":{"PSNR":"30.83"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2203.14537","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}