Iterative Nearest Neighbors
作者：
Highlights：
• We propose the Iterative Nearest Neighbors (INN) representation and its refined variant.
• INN proves on par or better performance with MP and OMP on sparse signal recovery task.
• We derive INN-based dimensionality reduction and classification methods.
• We use face (AR), traffic sign (GTSRB), scene (Scene-15) and obj. class (Pascal VOC) benchmarks.
• INN proves performance on par with SR but running times closer to NN, for low dimensional data.
摘要
Highlights•We propose the Iterative Nearest Neighbors (INN) representation and its refined variant.•INN proves on par or better performance with MP and OMP on sparse signal recovery task.•We derive INN-based dimensionality reduction and classification methods.•We use face (AR), traffic sign (GTSRB), scene (Scene-15) and obj. class (Pascal VOC) benchmarks.•INN proves performance on par with SR but running times closer to NN, for low dimensional data.
论文关键词：Iterative Nearest Neighbors,Least squares,Sparse representation,Collaborative representation,Classification,Dimensionality reduction
论文评审过程：Received 5 May 2013, Revised 6 June 2014, Accepted 10 July 2014, Available online 21 July 2014.
论文官网地址：https://doi.org/10.1016/j.patcog.2014.07.011
