🤖 AI Summary
To address poor generalization and severe content interference in implicit degradation modeling for blind image super-resolution (SR), this paper proposes a content-decoupled, negative-free contrastive learning framework. Methodologically, it introduces negative-free contrastive learning (NNCL) into degradation representation learning for blind SR—the first such application—designs a cyclic shift sampling strategy to explicitly decouple content and degradation features at the data level, and develops a detail-aware lightweight degradation adaptation module to significantly reduce model complexity. Evaluated on both synthetic and real-world degradation datasets, the method achieves state-of-the-art quantitative metrics and superior visual quality with substantially fewer parameters and lower computational overhead. These results validate the feasibility of an efficient, practical, and lightweight blind SR modeling paradigm.
📝 Abstract
Implicit degradation modeling-based blind super-resolution (SR) has attracted more increasing attention in the community due to its excellent generalization to complex degradation scenarios and wide application range. How to extract more discriminative degradation representations and fully adapt them to specific image features is the key to this task. In this paper, we propose a new Content-decoupled Contrastive Learning-based blind image super-resolution (CdCL) framework following the typical blind SR pipeline. This framework introduces negative-free contrastive learning technique for the first time to model the implicit degradation representation, in which a new cyclic shift sampling strategy is designed to ensure decoupling between content features and degradation features from the data perspective, thereby improving the purity and discriminability of the learned implicit degradation space. In addition, we propose a detail-aware implicit degradation adapting module that can better adapt degradation representations to specific LR features by enhancing the basic adaptation unit's perception of image details, significantly reducing the overall SR model complexity. Extensive experiments on synthetic and real data show that our method achieves highly competitive quantitative and qualitative results in various degradation settings while obviously reducing parameters and computational costs, validating the feasibility of designing practical and lightweight blind SR tools.