DACESR: Degradation-Aware Conditional Embedding for Real-World Image Super-Resolution
- 12 March 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 35 (10577149) , 2997-3008
Abstract
Multimodal large models have shown excellent ability in addressing image super-resolution in real-world scenarios by leveraging language class as condition information, yet their abilities in degraded images remain limited. In this paper, we first revisit the capabilities of the Recognize Anything Model (RAM) for degraded images by calculating text similarity. We find that directly using contrastive learning to fine-tune RAM in the degraded space is difficult to achieve acceptable results. To address this issue, we employ a degradation selection strategy to propose a Real Embedding Extractor (REE), which achieves significant recognition performance gain on degraded image content through contrastive learning. Furthermore, we use a Conditional Feature Modulator (CFM) to incorporate the high-level information of REE for a powerful Mamba-based network, which can leverage effective pixel information to restore image textures and produce visually pleasing results. Extensive experiments demonstrate that the REE can effectively help image super-resolution networks balance fidelity and perceptual quality, highlighting the great potential of Mamba in real-world applications. The source code of this work will be made publicly available at: https://github.com/nathan66666/DACESR.git.Keywords
All Related Versions
Funding Information
- Graduate Education Reform Project of Henan Province (2023SJGLX037Y)
- Research Project of Humanities and Social Sciences of the Ministry of Education, China, (24YJAZH075)
- Research Project of Humanities and Social Sciences of Henan Province
This publication has 48 references indexed in Scilit:
- MambaIR: A Simple Baseline for Image Restoration with State-Space Model2024
- Exploiting Diffusion Prior for Real-World Image Super-ResolutionInternational Journal of Computer Vision, 2024
- Knowledge Distillation Based Fast Degradation Estimator For Blind Super-resolution2023
- Long Movie Clip Classification with State-Space Video Models2022
- Efficient and Degradation-Adaptive Network for Real-World Image Super-Resolution2022
- Conditional Sequential Modulation for Efficient Global Image Retouching2020
- Component Divide-and-Conquer for Real-World Image Super-Resolution2020
- Residual Feature Distillation Network for Lightweight Image Super-Resolution2020
- Accelerating the Super-Resolution Convolutional Neural Network2016
- Microsoft COCO: Common Objects in Context2014