Iterative ensemble normalized cuts

作者：

Highlights：

• We provide a novel interpretation of the gap-view concept in NCut.

• We improve NCut by assembling out-of-sample extensions of multiple training sets.

• We speed up the clustering method by employing an iterative algorithm.

• Our proposed algorithm performs better than the state-of-the-art algorithms.

摘要

Highlights•We provide a novel interpretation of the gap-view concept in NCut.•We improve NCut by assembling out-of-sample extensions of multiple training sets.•We speed up the clustering method by employing an iterative algorithm.•Our proposed algorithm performs better than the state-of-the-art algorithms.

论文关键词：Iterative ensemble NCut,Gap-normalized distance,Spectral clustering,Image segmentation

论文评审过程：Received 2 December 2014, Revised 7 October 2015, Accepted 28 October 2015, Available online 10 November 2015, Version of Record 24 December 2015.

论文官网地址：https://doi.org/10.1016/j.patcog.2015.10.019