Async-RED: A Provably Convergent Asynchronous Block Parallel Stochastic Method using Deep Denoising Priors
Async-RED: A Provably Convergent Asynchronous Block Parallel Stochastic Method using Deep Denoising Priors
复制标题
DOI：
--
复制
发表时间：
2020-10
期刊：
影响因子：
--
通讯作者：
Yu Sun;Jiaming Liu;Yiran Sun;B. Wohlberg;U. Kamilov
中科院分区：
文献类型：
--
作者：
Yu Sun;Jiaming Liu;Yiran Sun;B. Wohlberg;U. Kamilov
去噪正则化（RED）是最近发展起来的一种框架，它通过将高级去噪器作为图像先验来求解逆问题。最近的研究表明，当与预训练的深度去噪器结合使用时，它具有最先进的性能。然而，目前的RED算法不足以在多核系统上进行并行处理。我们通过提出一种新的异步RED （ASYNC-RED）算法来解决这个问题，该算法支持异步并行处理数据，使其在处理大规模逆问题时明显快于串行算法。通过在每次迭代中使用随机的测量子集，进一步降低了ASYNC-RED的计算复杂度。我们通过在数据保真度和去噪的明确假设下建立算法的收敛性，对该算法进行了完整的理论分析。我们使用预训练的深度去噪器作为先验验证ASYNC-RED在图像恢复上的有效性。
Regularization by denoising (RED) is a recently developed framework for solving inverse problems by integrating advanced denoisers as image priors. Recent work has shown its state-of-the-art performance when combined with pre-trained deep denoisers. However, current RED algorithms are inadequate for parallel processing on multicore systems. We address this issue by proposing a new asynchronous RED (ASYNC-RED) algorithm that enables asynchronous parallel processing of data, making it significantly faster than its serial counterparts for large-scale inverse problems. The computational complexity of ASYNC-RED is further reduced by using a random subset of measurements at every iteration. We present complete theoretical analysis of the algorithm by establishing its convergence under explicit assumptions on the data-fidelity and the denoiser. We validate ASYNC-RED on image recovery using pre-trained deep denoisers as priors.