Flexible constrained sparsity preserving embedding

作者：

Highlights：

• Two non-linear semi-supervised embeddings are proposed.

• These methods elegantly integrate sparsity preserving and constrained embedding.

• The second framework provides a non-linear embedding and its out-of-sample extension.

• Classification performance after embedding is assessed on eight image datasets.

• KNN and SVM classifiers are used after getting the embedding.

• Experimental results on eight public image datasets show the outperformance of the methods.

摘要

Highlights•Two non-linear semi-supervised embeddings are proposed.•These methods elegantly integrate sparsity preserving and constrained embedding.•The second framework provides a non-linear embedding and its out-of-sample extension.•Classification performance after embedding is assessed on eight image datasets.•KNN and SVM classifiers are used after getting the embedding.•Experimental results on eight public image datasets show the outperformance of the methods.

论文关键词：Constrained embedding,Sparsity preserving projections,Flexible manifold embedding,Semi-supervised learning,Out-of-sample problem

论文评审过程：Received 29 February 2016, Revised 15 June 2016, Accepted 28 June 2016, Available online 30 June 2016, Version of Record 21 July 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.06.027