Robust Subspace Clustering with Compressed Data
arXiv:1803.11305 · doi:10.1109/TIP.2019.2917857
Abstract
Dimension reduction is widely regarded as an effective way for decreasing the computation, storage and communication loads of data-driven intelligent systems, leading to a growing demand for statistical methods that allow analysis (e.g., clustering) of compressed data. We therefore study in this paper a novel problem called compressive robust subspace clustering, which is to perform robust subspace clustering with the compressed data, and which is generated by projecting the original high-dimensional data onto a lower-dimensional subspace chosen at random. Given only the compressed data and sensing matrix, the proposed method, row space pursuit (RSP), recovers the authentic row space that gives correct clustering results under certain conditions. Extensive experiments show that RSP is distinctly better than the competing methods, in terms of both clustering accuracy and computational efficiency.
References in corpus (2)
Cited by in corpus (12)
- Latent Complete Row Space Recovery for Multi-view Subspace Clustering
- Adaptive Attribute and Structure Subspace Clustering Network
- Convex Subspace Clustering by Adaptive Block Diagonal Representation
- Structured Graph Learning for Scalable Subspace Clustering: From Single-view to Multi-view
- Multilayer Collaborative Low-Rank Coding Network for Robust Deep Subspace Discovery
- Towards Clustering-friendly Representations: Subspace Clustering via Graph Filtering
- Task-Feature Collaborative Learning with Application to Personalized Attribute Prediction
- Discriminative Local Sparse Representation by Robust Adaptive Dictionary Pair Learning
- Flexible Auto-weighted Local-coordinate Concept Factorization: A Robust Framework for Unsupervised Clustering
- Convolutional Dictionary Pair Learning Network for Image Representation Learning
- Low Rank Regularization: A Review
- Multi-view Subspace Clustering via Partition Fusion