Search papers, labs, and topics across Lattice.
The paper introduces PRISM, a Proximal Scattered Atmosphere Reconstruction framework, to jointly estimate clean images and scattering variables for real-world image dehazing. This approach leverages a physically structured atmospheric scattering model to handle complex haze distribution and varying illumination. The model is trained using a novel online non-uniform haze synthesis pipeline and a Selective Self-distillation Adaptation scheme for unpaired real data, achieving state-of-the-art performance on real-world dehazing benchmarks.
Achieve state-of-the-art real-world image dehazing by jointly reconstructing the clear scene and scattering variables, even with non-uniform haze and complex lighting.
Real-world image dehazing (RID) aims to remove haze induced degradation from real scenes. This task remains challenging due to non-uniform haze distribution, spatially varying illumination from multiple light sources, and the scarcity of paired real hazy-clean data. In PRISM, we propose Proximal Scattered Atmosphere Reconstruction (PSAR), a physically structured framework that jointly reconstructs the clear scene and scattering variables under the atmospheric scattering model, thereby improving reliability in complex regions and mixed-light conditions. To bridge the synthetic-to-real gap, we design an online non-uniform haze synthesis pipeline and a Selective Self-distillation Adaptation scheme for unpaired real-world scenarios, which enables the model to selectively learn from high-quality perceptual targets while leveraging its intrinsic scattering understanding to audit residual haze and guide self-refinement. Extensive experiments on real-world benchmarks demonstrate that PRISM achieves state-of-the-art performance on RID tasks.
Tsinghua AI