# Multi-level Graph Subspace Contrastive Learning for Hyperspectral Image Clustering

arXiv: https://arxiv.org/abs/2404.05211 · HTML: https://mlsanity.com/paper/2404.05211/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 56 references that are in the map.
- First posted on arXiv 8 Apr 2024.

## Builds on (most cited first, 20 of 20)

- [Deep Graph Clustering via Dual Correlation Reduction](https://mlsanity.com/paper/2112.14772/) (2021)
- [Deep Fusion Clustering Network](https://mlsanity.com/paper/2012.09600/) (2020)
- [Hard Sample Aware Network for Contrastive Deep Graph Clustering](https://mlsanity.com/paper/2212.08665/) (2022)
- [Simple Contrastive Graph Clustering](https://mlsanity.com/paper/2205.07865/) (2022)
- [Dink-Net: Neural Clustering on Large Graphs](https://mlsanity.com/paper/2305.18405/) (2023)
- [Towards High-quality HDR Deghosting with Conditional Diffusion Models](https://mlsanity.com/paper/2311.00932/) (2023)
- [Global and Local Contrastive Self-Supervised Learning for Semantic Segmentation of HR Remote Sensing Images](https://mlsanity.com/paper/2106.10605/) (2021)
- [Multi-view Low-rank Sparse Subspace Clustering](https://mlsanity.com/paper/1708.08732/) (2017)
- [GaTector: A Unified Framework for Gaze Object Prediction](https://mlsanity.com/paper/2112.03549/) (2021)
- [Graph Convolutional Subspace Clustering: A Robust Subspace Clustering Framework for Hyperspectral Image](https://mlsanity.com/paper/2004.10476/) (2020)
- [A Novel Approach for Effective Multi-View Clustering with Information-Theoretic Perspective](https://mlsanity.com/paper/2309.13989/) (2023)
- [Deep Multi-View Subspace Clustering with Anchor Graph](https://mlsanity.com/paper/2305.06939/) (2023)
- [Contrastive Multi-view Subspace Clustering of Hyperspectral Images based on Graph Convolutional Networks](https://mlsanity.com/paper/2312.06068/) (2023)
- [Pixel-Superpixel Contrastive Learning and Pseudo-Label Correction for Hyperspectral Image Clustering](https://mlsanity.com/paper/2312.09630/) (2023)
- [Multi-class Support Vector Machine with Maximizing Minimum Margin](https://mlsanity.com/paper/2312.06578/) (2023)
- [Homophily-Related: Adaptive Hybrid Graph Filter for Multi-View Graph Clustering](https://mlsanity.com/paper/2401.02682/) (2024)
- [MCoCo: Multi-level Consistency Collaborative Multi-view Clustering](https://mlsanity.com/paper/2302.13339/) (2023)
- [Multi-view Semantic Consistency based Information Bottleneck for Clustering](https://mlsanity.com/paper/2303.00002/) (2023)
- [Mixture of personality improved Spiking actor network for efficient multi-agent cooperation](https://mlsanity.com/paper/2305.05898/) (2023)
- [Multiview Subspace Clustering of Hyperspectral Images based on Graph Convolutional Networks](https://mlsanity.com/paper/2403.01465/) (2024)

## Built on by (most cited first, 3 of 3)

- [Alignment Unlocks Complementarity: A Framework for Multiview Circuit Representation Learning](https://mlsanity.com/paper/2509.20968/) (2025)
- [GNN-based Path-aware multi-view Circuit Learning for Technology Mapping](https://mlsanity.com/paper/2601.14286/) (2026)
- [GTAC: A Generative Transformer for Approximate Circuits](https://mlsanity.com/paper/2601.19906/) (2025)

## Abstract (arXiv)

Hyperspectral image (HSI) clustering is a challenging task due to its high complexity. Despite subspace clustering shows impressive performance for HSI, traditional methods tend to ignore the global-local interaction in HSI data. In this study, we proposed a multi-level graph subspace contrastive learning (MLGSC) for HSI clustering. The model is divided into the following main parts. Graph convolution subspace construction: utilizing spectral and texture feautures to construct two graph convolution views. Local-global graph representation: local graph representations were obtained by step-by-step convolutions and a more representative global graph representation was obtained using an attention-based pooling strategy. Multi-level graph subspace contrastive learning: multi-level contrastive learning was conducted to obtain local-global joint graph representations, to improve the consistency of the positive samples between views, and to obtain more robust graph embeddings. Specifically, graph-level contrastive learning is used to better learn global representations of HSI data. Node-level intra-view and inter-view contrastive learning is designed to learn joint representations of local regions of HSI. The proposed model is evaluated on four popular HSI datasets: Indian Pines, Pavia University, Houston, and Xu Zhou. The overall accuracies are 97.75%, 99.96%, 92.28%, and 95.73%, which significantly outperforms the current state-of-the-art clustering methods.