Universal Compressed Image Restoration via Codec-Aware Conditioning with Reinforcement Learning
Changwoo Han, Hongil Kim, Donghyun Kim, Sung-Chang Lim, Seung-Won Jung
Abstract
We address the task of universal compressed image restoration, which involves recovering high-quality images degraded by a wide range of codecs and compression levels. While prior methods have made significant progress, they typically target specific degradation types and struggle to generalize across both traditional and learning-based codecs. To overcome this limitation, we propose a unified framework that leverages codec-aware conditioning and reinforcement learning-based fine-tuning. Specifically, we introduce a conditioning module that encodes both codec type and compression level, enabling the restoration network to adapt its behavior to diverse degradation settings. To further improve generalization, we incorporate reward-based objectives during fine-tuning, providing complementary signals that enhance training across both conventional and learned compression schemes. Experimental results demonstrate the effectiveness of our method in restoring images across a wide range of compression artifacts and scenarios.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
Builds on12
- Learning Transferable Visual Models From Natural Language SupervisionAlec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh et al.ICML 2021 · 47,906 citations
- Gradient Surgery for Multi-Task LearningTianhe Yu, Saurabh Kumar, Abhishek Gupta, Sergey Levine et al.NeurIPS 2020 · 2,261 citations
- High-Fidelity Generative Image CompressionFabian Mentzer, George Toderici, Michael Tschannen, Eirikur AgustssonNeurIPS 2020 · 675 citations
- PromptIR: Prompting for All-in-One Image RestorationVaishnav Potlapalli, Syed Waqas Zamir, Salman H. Khan, Fahad Shahbaz KhanNeurIPS 2023 · 386 citations
- All-In-One Image Restoration for Unknown CorruptionBoyun Li, Xiao Liu, Peng Hu, Zhongqin Wu et al.CVPR 2022 · 338 citations
Related papers
- Variable Rate Deep Image Compression With a Conditional AutoencoderYoojin Choi, Mostafa El-Khamy, Jungwon LeeICCV 2019 · 265 citations
- UniRes: Universal Image Restoration for Complex DegradationsMo Zhou, Keren Ye, Mauricio Delbracio, Peyman Milanfar et al.ICCV 2025
- UniFlowRestore: A General Video Restoration Framework via Flow Matching and Prompt GuidanceShuning Sun, Yu Zhang, Chen Wu, Dianjie Lu et al.ACM MM 2025 · 3 citations
- UniRestore: Unified Perceptual and Task-Oriented Image Restoration Model Using Diffusion PriorI-Hsiang Chen, Wei-Ting Chen, Yu-Wei Liu, Yuan-Chun Chiang et al.CVPR 2025
- Learning Weather-General and Weather-Specific Features for Image Restoration Under Multiple Adverse Weather ConditionsYurui Zhu, Tianyu Wang, Xueyang Fu, Xuanyu Yang et al.CVPR 2023
