{"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/data-upcycling-knowledge-distillation-for","title":"Data Upcycling Knowledge Distillation for Image Super-Resolution","arxiv_id":"2309.14162","date":"2023-09-25","proceeding":null,"authors":["Yun Zhang","Wei Li","Simiao Li","Hanting Chen","Zhijun Tu","Wenjia Wang","BingYi Jing","Shaohui Lin","Jie Hu"],"abstract":"Knowledge distillation (KD) compresses deep neural networks by transferring task-related knowledge from cumbersome pre-trained teacher models to compact student models. However, current KD methods for super-resolution (SR) networks overlook the nature of SR task that the outputs of the teacher model are noisy approximations to the ground-truth distribution of high-quality images (GT), which shades the teacher model's knowledge to result in limited KD effects. To utilize the teacher model beyond the GT upper-bound, we present the Data Upcycling Knowledge Distillation (DUKD), to transfer the teacher model's knowledge to the student model through the upcycled in-domain data derived from training data. Besides, we impose label consistency regularization to KD for SR by the paired invertible augmentations to improve the student model's performance and robustness. Comprehensive experiments demonstrate that the DUKD method significantly outperforms previous arts on several SR tasks.","url_abs":"https://arxiv.org/abs/2309.14162v4","url_pdf":"https://arxiv.org/pdf/2309.14162v4.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":"data-upcycling-knowledge-distillation-for","repo_url":"https://github.com/yun224/dukd","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"image-super-resolution","task_name":"Image Super-Resolution"},{"task_slug":"knowledge-distillation","task_name":"Knowledge Distillation"},{"task_slug":"quantization","task_name":"Quantization"},{"task_slug":"super-resolution","task_name":"Super-Resolution"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"knowledge-distillation","method_name":"Knowledge Distillation"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-super-resolution-on-set14-4x-upscaling","task":"Image Super-Resolution","dataset":"Set14 - 4x upscaling","model":"DUKD","rank_in_archive_order":50,"of":104,"metrics":{"PSNR":"28.80","SSIM":"0.7866"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-urban100-4x","task":"Image Super-Resolution","dataset":"Urban100 - 4x upscaling","model":"RCAN (DUKD)","rank_in_archive_order":32,"of":65,"metrics":{"PSNR":"26.62","SSIM":"0.802"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-urban100-4x","task":"Image Super-Resolution","dataset":"Urban100 - 4x upscaling","model":"EDSR (DUKD)","rank_in_archive_order":37,"of":65,"metrics":{"PSNR":"26.45","SSIM":"0.7963"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-urban100-4x","task":"Image Super-Resolution","dataset":"Urban100 - 4x upscaling","model":"SwinIR (DUKD)","rank_in_archive_order":38,"of":65,"metrics":{"PSNR":"26.43","SSIM":"0.7972"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2309.14162","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}