Zhang Youjian, Wang Chaoyue, Tao Dacheng
IEEE Trans Pattern Anal Mach Intell. 2023 Dec;45(12):15203-15218. doi: 10.1109/TPAMI.2023.3303450. Epub 2023 Nov 3.
Real-world dynamic scene deblurring has long been a challenging task since paired blurry-sharp training data is unavailable. Conventional Maximum A Posteriori estimation and deep learning-based deblurring methods are restricted by handcrafted priors and synthetic blurry-sharp training pairs respectively, thereby failing to generalize to real dynamic blurriness. To this end, we propose a Neural Maximum A Posteriori (NeurMAP) estimation framework for training neural networks to recover blind motion information and sharp content from unpaired data. The proposed NeruMAP consists of a motion estimation network and a deblurring network which are trained jointly to model the (re)blurring process (i.e. likelihood function). Meanwhile, the motion estimation network is trained to explore the motion information in images by applying implicit dynamic motion prior, and in return enforces the deblurring network training (i.e. providing sharp image prior). The proposed NeurMAP is an orthogonal approach to existing deblurring neural networks, and is the first framework that enables training image deblurring networks on unpaired datasets. Experiments demonstrate our superiority on both quantitative metrics and visual quality over State-of-the-art methods.
长期以来，真实世界动态场景去模糊一直是一项具有挑战性的任务，因为难以获得清晰与模糊图像配对的训练数据。传统的最大后验估计和基于深度学习的去模糊方法分别受到手工制作的先验和合成清晰与模糊训练对的限制，因此无法推广到真实的动态模糊情况。为此，我们提出了一种神经最大后验（NeurMAP）估计框架，用于训练神经网络从未配对的数据中恢复盲运动信息和清晰内容。所提出的NeurMAP由一个运动估计网络和一个去模糊网络组成，它们联合训练以对（重新）模糊过程（即似然函数）进行建模。同时，运动估计网络通过应用隐式动态运动先验来训练，以探索图像中的运动信息，进而强化去模糊网络的训练（即提供清晰图像先验）。所提出的NeurMAP是一种与现有去模糊神经网络正交的方法，并且是第一个能够在未配对数据集上训练图像去模糊网络的框架。实验表明，我们在定量指标和视觉质量方面均优于现有最先进的方法。
