A Physics-Informed Blur Learning Framework for Imaging Systems
This work addresses blur estimation for imaging applications, offering a novel method that reduces optimization complexity and eliminates the need for lens parameters, though it appears incremental in the context of existing PSF models.
The paper tackles the problem of accurate blur estimation for imaging systems by proposing a physics-informed PSF learning framework, which improves image quality in simulations and shows visual enhancements on real images compared to recent methods.
Accurate blur estimation is essential for high-performance imaging across various applications. Blur is typically represented by the point spread function (PSF). In this paper, we propose a physics-informed PSF learning framework for imaging systems, consisting of a simple calibration followed by a learning process. Our framework could achieve both high accuracy and universal applicability. Inspired by the Seidel PSF model for representing spatially varying PSF, we identify its limitations in optimization and introduce a novel wavefront-based PSF model accompanied by an optimization strategy, both reducing optimization complexity and improving estimation accuracy. Moreover, our wavefront-based PSF model is independent of lens parameters, eliminate the need for prior knowledge of the lens. To validate our approach, we compare it with recent PSF estimation methods (Degradation Transfer and Fast Two-step) through a deblurring task, where all the estimated PSFs are used to train state-of-the-art deblurring algorithms. Our approach demonstrates improvements in image quality in simulation and also showcases noticeable visual quality improvements on real captured images.