ProDehaze: Prompting Diffusion Models Toward Faithful Image Dehazing
This work addresses hallucination issues in image dehazing for computer vision applications, representing an incremental improvement over existing methods.
The paper tackles the problem of hallucination in image dehazing using pretrained diffusion models by proposing ProDehaze, which uses internal priors to guide the models, resulting in high-fidelity outputs with reduced color shifts as demonstrated on real-world datasets.
Recent approaches using large-scale pretrained diffusion models for image dehazing improve perceptual quality but often suffer from hallucination issues, producing unfaithful dehazed image to the original one. To mitigate this, we propose ProDehaze, a framework that employs internal image priors to direct external priors encoded in pretrained models. We introduce two types of \textit{selective} internal priors that prompt the model to concentrate on critical image areas: a Structure-Prompted Restorer in the latent space that emphasizes structure-rich regions, and a Haze-Aware Self-Correcting Refiner in the decoding process to align distributions between clearer input regions and the output. Extensive experiments on real-world datasets demonstrate that ProDehaze achieves high-fidelity results in image dehazing, particularly in reducing color shifts. Our code is at https://github.com/TianwenZhou/ProDehaze.