{"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/boosting-flow-based-generative-super","title":"Boosting Flow-based Generative Super-Resolution Models via Learned Prior","arxiv_id":"2403.10988","date":"2024-03-16","proceeding":"CVPR 2024 1","authors":["Li-Yuan Tsao","Yi-Chen Lo","Chia-Che Chang","Hao-Wei Chen","Roy Tseng","Chien Feng","Chun-Yi Lee"],"abstract":"Flow-based super-resolution (SR) models have demonstrated astonishing capabilities in generating high-quality images. However, these methods encounter several challenges during image generation, such as grid artifacts, exploding inverses, and suboptimal results due to a fixed sampling temperature. To overcome these issues, this work introduces a conditional learned prior to the inference phase of a flow-based SR model. This prior is a latent code predicted by our proposed latent module conditioned on the low-resolution image, which is then transformed by the flow model into an SR image. Our framework is designed to seamlessly integrate with any contemporary flow-based SR model without modifying its architecture or pre-trained weights. We evaluate the effectiveness of our proposed framework through extensive experiments and ablation analyses. The proposed framework successfully addresses all the inherent issues in flow-based SR models and enhances their performance in various SR scenarios. Our code is available at: https://github.com/liyuantsao/BFSR","url_abs":"https://arxiv.org/abs/2403.10988v3","url_pdf":"https://arxiv.org/pdf/2403.10988v3.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":"boosting-flow-based-generative-super","repo_url":"https://github.com/liyuantsao/BFSR","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"image-super-resolution","task_name":"Image Super-Resolution"},{"task_slug":"super-resolution","task_name":"Super-Resolution"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-super-resolution-on-div2k-val-4x","task":"Image Super-Resolution","dataset":"DIV2K val - 4x upscaling","model":"LINF-LP","rank_in_archive_order":5,"of":21,"metrics":{"LPIPS":"0.105","LRPSNR":"47.3","PSNR":"28.00","SSIM":"0.78"},"uses_additional_data":true},{"leaderboard":"/sota/image-super-resolution-on-div2k-val-4x","task":"Image Super-Resolution","dataset":"DIV2K val - 4x upscaling","model":"SRFlow-LP","rank_in_archive_order":6,"of":21,"metrics":{"LPIPS":"0.109","LRPSNR":"51.51","PSNR":"27.51","SSIM":"0.78"},"uses_additional_data":true}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2403.10988","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}