Scalable Deep Subspace Clustering Network

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Mrabah, Nairouz, Bouguessa, Mohamed, Sami, Sihem
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908731724267520
author Mrabah, Nairouz
Bouguessa, Mohamed
Sami, Sihem
author_facet Mrabah, Nairouz
Bouguessa, Mohamed
Sami, Sihem
contents Subspace clustering methods face inherent scalability limits due to the $O(n^3)$ cost (with $n$ denoting the number of data samples) of constructing full $n\times n$ affinities and performing spectral decomposition. While deep learning-based approaches improve feature extraction, they maintain this computational bottleneck through exhaustive pairwise similarity computations. We propose SDSNet (Scalable Deep Subspace Network), a deep subspace clustering framework that achieves $\mathcal{O}(n)$ complexity through (1) landmark-based approximation, avoiding full affinity matrices, (2) joint optimization of auto-encoder reconstruction with self-expression objectives, and (3) direct spectral clustering on factorized representations. The framework combines convolutional auto-encoders with subspace-preserving constraints. Experimental results demonstrate that SDSNet achieves comparable clustering quality to state-of-the-art methods with significantly improved computational efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2512_21434
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Scalable Deep Subspace Clustering Network
Mrabah, Nairouz
Bouguessa, Mohamed
Sami, Sihem
Computer Vision and Pattern Recognition
Machine Learning
I.2.6; I.5.3
Subspace clustering methods face inherent scalability limits due to the $O(n^3)$ cost (with $n$ denoting the number of data samples) of constructing full $n\times n$ affinities and performing spectral decomposition. While deep learning-based approaches improve feature extraction, they maintain this computational bottleneck through exhaustive pairwise similarity computations. We propose SDSNet (Scalable Deep Subspace Network), a deep subspace clustering framework that achieves $\mathcal{O}(n)$ complexity through (1) landmark-based approximation, avoiding full affinity matrices, (2) joint optimization of auto-encoder reconstruction with self-expression objectives, and (3) direct spectral clustering on factorized representations. The framework combines convolutional auto-encoders with subspace-preserving constraints. Experimental results demonstrate that SDSNet achieves comparable clustering quality to state-of-the-art methods with significantly improved computational efficiency.
title Scalable Deep Subspace Clustering Network
topic Computer Vision and Pattern Recognition
Machine Learning
I.2.6; I.5.3
url https://arxiv.org/abs/2512.21434