Super-resolution using neighbourhood regression with local structure prior
作者：
Highlights：
• A measure is proposed for extracting local structural similarity. For each patch, the local structure prior can yield a robust representation.
• An algorithm is developed for learning multiple local mappings. Different linear mappings are used simultaneously to reconstruct the desired image.
• A model is proposed that combines hierarchical similarity and local structure priors. This model yields more reasonable and reliable results than the existing methods based on clustering or regression alone.
摘要
•A measure is proposed for extracting local structural similarity. For each patch, the local structure prior can yield a robust representation.•An algorithm is developed for learning multiple local mappings. Different linear mappings are used simultaneously to reconstruct the desired image.•A model is proposed that combines hierarchical similarity and local structure priors. This model yields more reasonable and reliable results than the existing methods based on clustering or regression alone.
论文关键词：Super-resolution,Clustering,Regression,Structure prior
论文评审过程：Received 9 June 2018, Revised 28 September 2018, Accepted 11 December 2018, Available online 18 December 2018, Version of Record 21 December 2018.
论文官网地址：https://doi.org/10.1016/j.image.2018.12.006
