{"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/cdfi-compression-driven-network-design-for","title":"CDFI: Compression-Driven Network Design for Frame Interpolation","arxiv_id":"2103.10559","date":"2021-03-18","proceeding":"CVPR 2021 1","authors":["Tianyu Ding","Luming Liang","Zhihui Zhu","Ilya Zharkov"],"abstract":"DNN-based frame interpolation--that generates the intermediate frames given two consecutive frames--typically relies on heavy model architectures with a huge number of features, preventing them from being deployed on systems with limited resources, e.g., mobile devices. We propose a compression-driven network design for frame interpolation (CDFI), that leverages model pruning through sparsity-inducing optimization to significantly reduce the model size while achieving superior performance. Concretely, we first compress the recently proposed AdaCoF model and show that a 10X compressed AdaCoF performs similarly as its original counterpart; then we further improve this compressed model by introducing a multi-resolution warping module, which boosts visual consistencies with multi-level details. As a consequence, we achieve a significant performance gain with only a quarter in size compared with the original AdaCoF. Moreover, our model performs favorably against other state-of-the-arts in a broad range of datasets. Finally, the proposed compression-driven framework is generic and can be easily transferred to other DNN-based frame interpolation algorithm. Our source code is available at https://github.com/tding1/CDFI.","url_abs":"https://arxiv.org/abs/2103.10559v2","url_pdf":"https://arxiv.org/pdf/2103.10559v2.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":"cdfi-compression-driven-network-design-for","repo_url":"https://github.com/tding1/CDFI","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"video-frame-interpolation","task_name":"Video Frame Interpolation"}],"methods":[{"method_slug":"deformable-convolution","method_name":"Deformable Convolution"},{"method_slug":"pruning","method_name":"Pruning"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/video-frame-interpolation-on-msu-video-frame","task":"Video Frame Interpolation","dataset":"MSU Video Frame Interpolation","model":"CDFI","rank_in_archive_order":17,"of":24,"metrics":{"LPIPS":"0.051","MS-SSIM":"0.926","PSNR":"26.99","SSIM":"0.908","VMAF":"61.72"},"uses_additional_data":false},{"leaderboard":"/sota/video-frame-interpolation-on-middlebury","task":"Video Frame Interpolation","dataset":"Middlebury","model":"CDFI","rank_in_archive_order":11,"of":11,"metrics":{"LPIPS":"0.007","PSNR":"37.14","SSIM":"0.966"},"uses_additional_data":false},{"leaderboard":"/sota/video-frame-interpolation-on-ucf101-1","task":"Video Frame Interpolation","dataset":"UCF101","model":"CDFI","rank_in_archive_order":13,"of":19,"metrics":{"LPIPS":"0.015","PSNR":"35.21"},"uses_additional_data":false},{"leaderboard":"/sota/video-frame-interpolation-on-vimeo90k","task":"Video Frame Interpolation","dataset":"Vimeo90K","model":"CDFI","rank_in_archive_order":14,"of":23,"metrics":{"LPIPS":"0.010","PSNR":"35.17"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2103.10559","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}