Abstract:Constructing an informative and discriminative graph plays an important role in the graph based semi-supervised learning methods. Among these graph construction methods, low-rank representation based graph, which calculates the edge weights of both labeled and unlabeled samples as the low-rank representation LRR coefficients, has shown excellent performance in semi-supervised learning. In this paper, we additionally impose twofold constraints local affinity and distant repulsion on the LRR graph. The improved model, termed structure preserving LRR SPLRR, can preserve the local geometrical structure but without distorting the distant repulsion property. Experiments are taken on three widely used face data sets to investigate the performance of SPLRR and the results show that it is superior to some state-of-the-art semi-supervised graphs.

Structure Preserving Low-Rank Representation for Semi-supervised Face Recognition.