Efficient deep learning of image denoising using patch complexity local divide and deep conquer

作者：

Highlights：

• Pattern complexity can be used to divide a task and improve learning efficiency.

• Weighted mixtures of locally-learned networks boost generalization performance.

• Training of networks is completely and naturally parallelizable by our strategy.

• Adaptive ensembles of local small-scale networks beat a single large-scale one.

摘要

•Pattern complexity can be used to divide a task and improve learning efficiency.•Weighted mixtures of locally-learned networks boost generalization performance.•Training of networks is completely and naturally parallelizable by our strategy.•Adaptive ensembles of local small-scale networks beat a single large-scale one.

论文关键词：Image denoising,Deep neural networks,Stacked denoising autoencoders,Divide and conquer,Local experts,Image patch complexity,Ensemble selection,Image patch classification

论文评审过程：Received 13 March 2018, Revised 26 December 2018, Accepted 15 June 2019, Available online 17 June 2019, Version of Record 2 September 2019.

论文官网地址：https://doi.org/10.1016/j.patcog.2019.06.011