Search papers, labs, and topics across Lattice.
To overcome the severe scarcity of polarized training data when handling complex real-world degradations, the authors formulate an all-in-one polarization image restoration framework grounded in normalized Stokes representations. The system decouples inputs into a dual-branch architecture, employing a mixture-of-experts general image restoration prior on the intensity stream that adaptively distills recovery knowledge into the polarization stream via cross-domain feature transforms. Evaluated on public datasets and a newly established composite-degradation benchmark, this architecture demonstrates that pre-trained conventional vision priors can effectively resolve coupled physical degradations in low-resource polarization domains.
Polarization vision can bypass its chronic data-scarcity bottleneck by decoupling Stokes parameters and borrowing pretrained RGB restoration priors via cross-domain feature distillation.
Polarization imaging captures distinctive surface and geometric cues that benefit a wide range of vision tasks. However, real-world polarization acquisition is often affected by multiple coupled degradations, making image restoration essential for practical polarization vision. Existing methods are largely tailored to specific degradations and remain constrained by the limited scale and quality of polarization data. To address these limitations, we develop an all-in-one polarization restoration framework for diverse and composite degradations. We first study the impact of different polarization representations on restoration performance and identify the normalized Stokes representation as an effective choice for separating intensity and polarization information. Accordingly, we devise a dual-branch architecture that separates intensity and polarization modeling. To overcome the limitations of polarization-specific training, the intensity branch leverages pretrained general restoration priors and a mixture-of-experts extension for composite degradations, while its restoration knowledge is adaptively distilled into the symmetric polarization branch via a cross-domain feature transform. In addition, we establish a composite-degradation polarization benchmark to support all-in-one restoration research. Extensive experiments on public datasets and our proposed benchmark demonstrate the effectiveness of the proposed method.
