Abstract: A new subspace structured low rank regularity is defined. Combining with the subspace Structured Sparse Subspace Clustering method（SSSC）, a new unified optimization model is given. The new model uses the estimated cluster membership of the samples and the affinity of the samples to guide each other so that the affinity possesses both discrimination and coherence and the cluster membership have coherence property. The discrimination of the affinity tends to segment data from different subspaces into different clusters while the coherence tends to group data from the same subspace into the same cluster. Experiments show that the proposed method outperforms the state-of-the-art two-stage methods and the SSSC method.

Key words: subspace clustering, subspace structured low rank, affinity, discrimination, coherence

摘要： 通过定义子空间结构化低秩正则项，将其与子空间结构化稀疏子空间聚类模型相结合，给出一个新的统一优化模型。新模型利用数据的类别属性和相似度互相引导，使得相似度具有判别性和一致性，类别属性具有一致性。相似度的判别性有利于将不同子空间的数据分为不同类，而一致性有利于将同一子空间的数据聚为一类。大量实验表明提出的方法优于一些典型的两步法和子空间结构化稀疏子空间聚类模型。

关键词: 子空间聚类, 子空间结构化低秩, 相似度, 判别性, 一致性