Bilevel Layer-Positioning LoRA for Real Image Dehazing
This addresses adaptation challenges in diverse real haze scenes for computer vision applications, representing an incremental improvement over existing learning-based methods.
The paper tackles real image dehazing by proposing a haze-to-clear text-directed loss using CLIP for semantic alignment and a Bilevel Layer-positioning LoRA strategy to adapt critical network layers, achieving state-of-the-art results on multiple real-world benchmarks.
Learning-based real image dehazing methods have achieved notable progress, yet they still face adaptation challenges in diverse real haze scenes. These challenges mainly stem from the lack of effective unsupervised mechanisms for unlabeled data and the heavy cost of full model fine-tuning. To address these challenges, we propose the haze-to-clear text-directed loss that leverages CLIP's cross-modal capabilities to reformulate real image dehazing as a semantic alignment problem in latent space, thereby providing explicit unsupervised cross-modal guidance in the absence of reference images. Furthermore, we introduce the Bilevel Layer-positioning LoRA (BiLaLoRA) strategy, which learns both the LoRA parameters and automatically search the injection layers, enabling targeted adaptation of critical network layers. Extensive experiments demonstrate our superiority against state-of-the-art methods on multiple real-world dehazing benchmarks. The code is publicly available at https://github.com/YanZhang-zy/BiLaLoRA.