Leveraging Phase Information to Boost Unrolled Network Learning for Image Deblurring
2026-06-30 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionArtificial Intelligence
AI summaryⓘ
The authors propose a new way to fix blurry images by separating the image into amplitude and phase parts, focusing on accurately recovering the phase to get clearer details. They create special estimators that help guess these parts from noisy, blurry pictures and then use an iterative algorithm to restore the sharp image. Their method, called UPADNet, learns key settings from example images during training, making each step of the process adjustable and trainable. Tests on popular datasets show that UPADNet does better than existing methods, especially when images are very noisy or there is little training data available.
image deblurringamplitude and phase decompositionLMMSE estimatoriterative optimizationalgorithm unrollingdeep learningimage restorationphase estimationtraining datasetnoise robustness
Authors
Samira Malek, Haichuan Zhang, Chul Lee, Vishal Monga
Abstract
While most image deblurring techniques directly restore the spatial image variable, we propose an amplitude and phase decomposition recognizing the importance of accurate phase estimation in recovering sharp image details. To that end, we first develop novel linear minimum mean squared (LMMSE) estimators of the amplitude and phase of the blurred, noisy image observation. An iterative optimization algorithm follows that recovers the sharp image using the aforementioned LMMSE estimators. Finally, matrix parameters that are statistically determined and fixed in the iterative algorithm are now learned using a training dataset of clean and degraded observations. Our deblurring engine is dubbed UPADNet (Unrolled Phase and Amplitude Decomposition Network), such that each iteration of the underlying phase and amplitude recovery algorithm is parameterized and trained end-to-end. Experiments over benchmark evaluation datasets such as GoPro, RealBlur and COCO datasets confirm that UPADNet outperforms state of the art deep networks including those based on algorithm unrolling in the image domain. The benefits of UPADNet are even more pronounced in high noise and limited training data regimes.