Beyond Pixels: Text Enhances Generalization in Real-World Image Restoration
This addresses generalization challenges in real-world image restoration for computer vision applications, representing an incremental improvement over existing diffusion-based methods.
The paper tackles the problem of diffusion-based image restoration models failing on out-of-distribution real-world data due to 'generative capability deactivation', and proposes using text as an auxiliary invariant representation to reactivate these capabilities, resulting in significantly enhanced generalization abilities as demonstrated through extensive experiments.
Generalization has long been a central challenge in real-world image restoration. While recent diffusion-based restoration methods, which leverage generative priors from text-to-image models, have made progress in recovering more realistic details, they still encounter "generative capability deactivation" when applied to out-of-distribution real-world data. To address this, we propose using text as an auxiliary invariant representation to reactivate the generative capabilities of these models. We begin by identifying two key properties of text input: richness and relevance, and examine their respective influence on model performance. Building on these insights, we introduce Res-Captioner, a module that generates enhanced textual descriptions tailored to image content and degradation levels, effectively mitigating response failures. Additionally, we present RealIR, a new benchmark designed to capture diverse real-world scenarios. Extensive experiments demonstrate that Res-Captioner significantly enhances the generalization abilities of diffusion-based restoration models, while remaining fully plug-and-play.