UniPolar restores images and polarization info amid mixed distortions
UniPolar: Joint Image and Polarimetric Restoration under Compound Degradations via Physical Consistency Learning
Computer Vision and Pattern Recognition
Summary
Polarimetric images show how light waves are oriented, which helps in many scientific and technical fields. But images from sensors can be blurry or noisy due to device issues and environmental effects at the same time. The authors introduce UniPolar, a method that fixes these images and recovers important polarization details by learning physical rules about the light's behavior. This approach improves both the quality of the image and the accuracy of the recovered polarization data.
What this means in practice
- •For imaging system engineers: Improve the reconstruction accuracy of polarimetric images distorted by both hardware defects and environmental noise using a unified restoration approach.
- •For industrial inspection teams: Obtain clearer and more reliable polarization-based images for quality control under challenging imaging conditions.
Authors
Yang Jin, Guangyu Guo, Linghao Shen, Binglu Wang
Abstract
Polarimetric image restoration aims to recover polarization angle images and reliable polarimetric quantities from degraded observations. Beyond visual fidelity, it also requires preserving the physical relationships underlying reliable polarimetric estimation. However, existing methods typically address either device- or environment-induced degradations in isolation, limiting their applicability to color polarization filter array (CPFA) imaging, where both often coexist. Such compound degradations jointly perturb image intensities and cross angle polarization differences. Consequently, improved image fidelity alone does not necessarily ensure accurate Stokes, DoLP, and AoLP recovery. To address this challenge, we propose UniPolar, a coupled Image Stokes framework for joint image polarimetric restoration from compound-degraded CPFA observations. Specifically, UniPolar employs two complementary streams to reconstruct four polarization-angle images and directly predict Stokes difference components. To coordinate these streams, Degradation Conditioned Cross Domain Interaction combines Dual Domain Frequency Interaction for frequency selective evidence exchange with Degradation Conditioned Modulation for condition adaptive feature correction. Furthermore, at the output level, Image Stokes Physical Consistency Learning aligns image derived and directly predicted Stokes components while regularizing angular fidelity and the physical feasibility of DoLP. On the compound-degradation benchmark, UniPolar achieves average image and DoLP PSNR of 31.522 dB and 24.097 dB, respectively, with an average AoLP error of 16.833 degree. These results underscore the importance of jointly considering visual fidelity and polarimetric reliability, offering a physically grounded perspective on learning based restoration under compound degradations.