# Multi-view Deep Subspace Clustering Networks

arXiv: https://arxiv.org/abs/1908.01978 · HTML: https://mlsanity.com/paper/1908.01978/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Aug 2019.

## Built on by (most cited first, 24 of 24)

- [Deep Clustering: A Comprehensive Survey](https://mlsanity.com/paper/2210.04142/) (2022)
- [Multi-VAE: Learning Disentangled View-common and View-peculiar Visual Representations for Multi-view Clustering](https://mlsanity.com/paper/2106.11232/) (2021)
- [Self-Supervised Discriminative Feature Learning for Deep Multi-View Clustering](https://mlsanity.com/paper/2103.15069/) (2021)
- [Deep Incomplete Multi-view Clustering with Cross-view Partial Sample and Prototype Alignment](https://mlsanity.com/paper/2303.15689/) (2023)
- [Pseudo-supervised Deep Subspace Clustering](https://mlsanity.com/paper/2104.03531/) (2021)
- [SPACE-2: Tree-Structured Semi-Supervised Contrastive Pre-training for Task-Oriented Dialog Understanding](https://mlsanity.com/paper/2209.06638/) (2022)
- [DealMVC: Dual Contrastive Calibration for Multi-view Clustering](https://mlsanity.com/paper/2308.09000/) (2023)
- [Self-Supervised Information Bottleneck for Deep Multi-View Subspace Clustering](https://mlsanity.com/paper/2204.12496/) (2022)
- [Deep Multi-View Subspace Clustering with Anchor Graph](https://mlsanity.com/paper/2305.06939/) (2023)
- [On the Effects of Self-supervision and Contrastive Alignment in Deep Multi-view Clustering](https://mlsanity.com/paper/2303.09877/) (2023)
- [Scalable Incomplete Multi-View Clustering with Structure Alignment](https://mlsanity.com/paper/2308.16541/) (2023)
- [Imputation-free and Alignment-free: Incomplete Multi-view Clustering Driven by Consensus Semantic Learning](https://mlsanity.com/paper/2505.11182/) (2025)
- [Fairness-aware Multi-view Clustering](https://mlsanity.com/paper/2302.05788/) (2023)
- [MetaViewer: Towards A Unified Multi-View Representation](https://mlsanity.com/paper/2303.06329/) (2023)
- [Subspace-Contrastive Multi-View Clustering](https://mlsanity.com/paper/2210.06795/) (2022)
- [Multi-Perspective Anomaly Detection](https://mlsanity.com/paper/2105.09903/) (2021)
- [Attentive Multi-View Deep Subspace Clustering Net](https://mlsanity.com/paper/2112.12506/) (2021)
- [Fine-grained Graph Learning for Multi-view Subspace Clustering](https://mlsanity.com/paper/2201.04604/) (2022)
- [Explicit View-labels Matter: A Multifacet Complementarity Study of Multi-view Clustering](https://mlsanity.com/paper/2205.02507/) (2022)
- [Multi-View Clustering from the Perspective of Mutual Information](https://mlsanity.com/paper/2302.08743/) (2023)
- [Self-supervised Multi-view Clustering in Computer Vision: A Survey](https://mlsanity.com/paper/2309.09473/) (2023)
- [Information Fusion: Scaling Subspace-Driven Approaches](https://mlsanity.com/paper/2204.12035/) (2022)
- [Federated Deep Subspace Clustering](https://mlsanity.com/paper/2501.00230/) (2024)
- [Personalized Federated Heat-Kernel Enhanced Multi-View Clustering via Advanced Tensor Decomposition Techniques](https://mlsanity.com/paper/2509.16101/) (2025)

## Abstract (arXiv)

Multi-view subspace clustering aims to discover the inherent structure of data by fusing multiple views of complementary information. Most existing methods first extract multiple types of handcrafted features and then learn a joint affinity matrix for clustering. The disadvantage of this approach lies in two aspects: 1) multi-view relations are not embedded into feature learning, and 2) the end-to-end learning manner of deep learning is not suitable for multi-view clustering. Even when deep features have been extracted, it is a nontrivial problem to choose a proper backbone for clustering on different datasets. To address these issues, we propose the Multi-view Deep Subspace Clustering Networks (MvDSCN), which learns a multi-view self-representation matrix in an end-to-end manner. The MvDSCN consists of two sub-networks, ie, a diversity network (Dnet) and a universality network (Unet). A latent space is built using deep convolutional autoencoders, and a self-representation matrix is learned in the latent space using a fully connected layer. Dnet learns view-specific self-representation matrices, whereas Unet learns a common self-representation matrix for all views. To exploit the complementarity of multi-view representations, the Hilbert--Schmidt independence criterion (HSIC) is introduced as a diversity regularizer that captures the nonlinear, high-order inter-view relations. Because different views share the same label space, the self-representation matrices of each view are aligned to the common one by universality regularization. The MvDSCN also unifies multiple backbones to boost clustering performance and avoid the need for model selection. Experiments demonstrate the superiority of the MvDSCN.