# Unsupervised CD in satellite image time series by contrastive learning and feature tracking

arXiv: https://arxiv.org/abs/2304.11375 · HTML: https://mlsanity.com/paper/2304.11375/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 58 references that are in the map.
- First posted on arXiv 22 Apr 2023.

## Builds on (most cited first, 18 of 18)

- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Generative Adversarial Networks](https://mlsanity.com/paper/2203.00667/) (2022)
- [Supervised Contrastive Learning](https://mlsanity.com/paper/2004.11362/) (2020)
- [Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting](https://mlsanity.com/paper/1506.04214/) (2015)
- [Self-supervised Learning: Generative or Contrastive](https://mlsanity.com/paper/2006.08218/) (2020)
- [Road Extraction by Deep Residual U-Net](https://mlsanity.com/paper/1711.10684/) (2017)
- [Fully Convolutional Siamese Networks for Change Detection](https://mlsanity.com/paper/1810.08462/) (2018)
- [Space-Time Correspondence as a Contrastive Random Walk](https://mlsanity.com/paper/2006.14613/) (2020)
- [LogME: Practical Assessment of Pre-trained Models for Transfer Learning](https://mlsanity.com/paper/2102.11005/) (2021)
- [The Multi-Temporal Urban Development SpaceNet Dataset](https://mlsanity.com/paper/2102.04420/) (2021)
- [Learning Spectral-Spatial-Temporal Features via a Recurrent Convolutional Neural Network for Change Detection in Multispectral Imagery](https://mlsanity.com/paper/1803.02642/) (2018)
- [Bi-Temporal Semantic Reasoning for the Semantic Change Detection in HR Remote Sensing Images](https://mlsanity.com/paper/2108.06103/) (2021)
- [Dense Unsupervised Learning for Video Segmentation](https://mlsanity.com/paper/2111.06265/) (2021)
- [Self-supervised Change Detection in Multi-view Remote Sensing Images](https://mlsanity.com/paper/2103.05969/) (2021)
- [Self-supervised pre-training enhances change detection in Sentinel-2 imagery](https://mlsanity.com/paper/2101.08122/) (2021)
- [Code-Aligned Autoencoders for Unsupervised Change Detection in Multimodal Remote Sensing Images](https://mlsanity.com/paper/2004.07011/) (2020)
- [Multi-Temporal Scene Classification and Scene Change Detection with Correlation based Fusion](https://mlsanity.com/paper/2006.02176/) (2020)
- [Unsupervised Self-training Algorithm Based on Deep Learning for Optical Aerial Images Change Detection](https://mlsanity.com/paper/2010.07469/) (2020)

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

- [Deep Learning for Satellite Image Time Series Analysis: A Review](https://mlsanity.com/paper/2404.03936/) (2024)
- [Multi-temporal Calving Front Segmentation](https://mlsanity.com/paper/2512.11560/) (2025)
- [Building Change Detection in Earthquake: A Multi-Scale Interaction Network and A Change Detection Dataset](https://mlsanity.com/paper/2606.10329/) (2026)

## Abstract (arXiv)

While unsupervised change detection using contrastive learning has been significantly improved the performance of literature techniques, at present, it only focuses on the bi-temporal change detection scenario. Previous state-of-the-art models for image time-series change detection often use features obtained by learning for clustering or training a model from scratch using pseudo labels tailored to each scene. However, these approaches fail to exploit the spatial-temporal information of image time-series or generalize to unseen scenarios. In this work, we propose a two-stage approach to unsupervised change detection in satellite image time-series using contrastive learning with feature tracking. By deriving pseudo labels from pre-trained models and using feature tracking to propagate them among the image time-series, we improve the consistency of our pseudo labels and address the challenges of seasonal changes in long-term remote sensing image time-series. We adopt the self-training algorithm with ConvLSTM on the obtained pseudo labels, where we first use supervised contrastive loss and contrastive random walks to further improve the feature correspondence in space-time. Then a fully connected layer is fine-tuned on the pre-trained multi-temporal features for generating the final change maps. Through comprehensive experiments on two datasets, we demonstrate consistent improvements in accuracy on fitting and inference scenarios.