Momentum-Net: Fast and Convergent Iterative Neural Network for Inverse Problems.
Momentum-Net: Fast and Convergent Iterative Neural Network for Inverse Problems.
复制标题
DOI：
10.1109/tpami.2020.3012955
复制
发表时间：
2023-04
影响因子：
23.6
通讯作者：
Fessler JA
中科院分区：
文献类型：
--
作者：
Chun IY;Huang Z;Lim H;Fessler JA
迭代神经网络（INN）正在迅速获得关注，用于解决成像，图像处理和计算机视觉中的逆问题。INN结合了联合收割机回归NN和基于迭代模型的图像重建（MBIR）算法，通常导致良好的泛化能力和优于现有MBIR优化模型的重建质量。本文提出了第一个快速收敛的INN架构，动量网络，通过推广一个块的MBIR算法，使用动量和优化回归神经网络。对于快速MBIR，Momentum-Net在外推模块中使用动量项，并通过使用优化器在每次迭代中使用非迭代MBIR模块，其中Momentum-Net的每次迭代由三个核心模块组成：图像细化，外推和MBIR。Momentum-Net保证了一般可微（非）凸MBIR函数（或数据拟合项）和凸可行集在两个无症状条件下收敛到不动点。为了考虑训练样本和测试样本之间的数据拟合变化，我们还提出了一种基于优化矩阵的“频谱扩展”的正则化参数选择方案。使用焦点堆栈和稀疏视图计算断层扫描的光场摄影的数值实验表明，给定相同的回归NN架构，Momentum-Net显着提高了MBIR的速度和准确性，超过了几个现有的INN;与最先进的MBIR方法相比，它显着提高了重建质量。
Iterative neural networks (INN) are rapidly gaining attention for solving inverse problems in imaging, image processing, and computer vision. INNs combine regression NNs and an iterative model-based image reconstruction (MBIR) algorithm, often leading to both good generalization capability and outperforming reconstruction quality over existing MBIR optimization models. This paper proposes the first fast and convergent INN architecture, Momentum-Net, by generalizing a block-wise MBIR algorithm that uses momentum and majorizers with regression NNs. For fast MBIR, Momentum-Net uses momentum terms in extrapolation modules, and noniterative MBIR modules at each iteration by using majorizers, where each iteration of Momentum-Net consists of three core modules: image refining, extrapolation, and MBIR. Momentum-Net guarantees convergence to a fixed-point for general differentiable (non)convex MBIR functions (or data-fit terms) and convex feasible sets, under two asymptomatic conditions. To consider data-fit variations across training and testing samples, we also propose a regularization parameter selection scheme based on the “spectral spread” of majorization matrices. Numerical experiments for light-field photography using a focal stack and sparse-view computational tomography demonstrate that, given identical regression NN architectures, Momentum-Net significantly improves MBIR speed and accuracy over several existing INNs; it significantly improves reconstruction quality compared to a state-of-the-art MBIR method in each application.