Research

My research sits at the intersection of machine learning and Earth observation, with the goal of making AI systems more capable, efficient, and accessible for monitoring and understanding our planet. My work spans three interconnected directions below.

Understanding Multi-Sensor EO Data

Modern Earth observation relies on a diverse fleet of sensors, each capturing complementary aspects of the Earth's surface. A central challenge is how to jointly interpret heterogeneous data streams that differ in modality, resolution, and acquisition geometry.

Representative Publications

Multimodal remote sensing change detection: An image matching perspective

H. Chen, C. Lan, J. Song, D. Ibañez, J. Xia, K. Schindler, N. Yokoya

ISPRS Journal of Photogrammetry and Remote Sensing, 2026

Paper Data

Fourier domain structural relationship analysis for unsupervised multimodal change detection

H. Chen, N. Yokoya, and M. Chini

ISPRS Journal of Photogrammetry and Remote Sensing, 2023

ESI Highly Cited Paper
Paper Code

Unsupervised multimodal change detection based on structural relationship graph representation learning

H. Chen, N. Yokoya, C. Wu, B. Du

IEEE Transactions on Geoscience and Remote Sensing, 2022

Paper Code

Designing EO-specialized Architecture

General-purpose vision architectures are not always well-suited to the unique characteristics of EO data. EO-specialized architectures are purpose-built for change detection, multitemporal analysis, and geospatial understanding. The goal is to close the gap between the structure of the problem and the inductive biases of the model.

Representative Publications

ChangeMamba: Remote sensing change detection with spatio-temporal state space model

H. Chen, J. Song, C. Han, J. Xia, and N. Yokoya

IEEE Transactions on Geoscience and Remote Sensing (TGRS), 2024

ESI Hot PaperESI Highly Cited Paper
Paper Code News

ObjFormer: Learning land-cover changes from paired OSM data and optical high-resolution imagery via object-guided Transformer

H. Chen, C. Lan, J. Song, C. Broni-Bediako, J. Xia, N. Yokoya

IEEE Transactions on Geoscience and Remote Sensing (TGRS), 2024

ESI Highly Cited Paper
Paper Code Datset

Unsupervised change detection in multitemporal VHR images based on deep kernel PCA convolutional mapping network

C. Wu, H. Chen, B. Du, and L. Zhang

IEEE Transactions on Cybernetics (TCYB), 2022

Paper Code Data

Lowering Training Cost for EO-AI

Supervised deep learning for remote sensing is bottlenecked by the scarcity and expense of high-quality labeled data. By lowering the barrier to entry for AI-powered Earth observation, this line of work aims to democratize access to intelligent geospatial analysis, particularly for regions and hazard types that are underrepresented in existing benchmarks.

Representative Publications

Bright: A globally distributed multimodal building damage assessment dataset with very-high-resolution for all-weather disaster response

H. Chen, J. Song, O. Dietrich, C. Broni-Bediako, W. Xuan, J. Wang, X. Shao, Y. Wei, J. Xia, C. Lan, K. Schindler and N. Yokoya

Earth System Science Data (ESSD), 2025

IEEE Data Fusion Contest 2025
Paper Code Data Contest

OpenEarthMap-SAR: A benchmark synthetic aperture radar dataset for global high-resolution land cover mapping

J. Xia, H. Chen, C. Broni-Bediako, Y. Wei, J. Song, N. Yokoya

IEEE Geoscience and Remote Sensing Magazine (GRSM), 2025

IEEE Data Fusion Contest 2025
Paper Code Data Contest

SynRS3D: A synthetic dataset for global 3D semantic understanding from monocular remote sensing imagery

J. Song, H. Chen, W. Xuan, J. Xia, N. Yokoya

The Thirty-eighth Annual Conference on Neural Information Processing Systems (NeurIPS), 2024

Spotlight Paper
Paper Code Data

Exchange means change: An unsupervised single-temporal change detection framework based on intra- and inter-image patch exchange

H. Chen, J. Song, C. Wu, B. Du, and N. Yokoya

ISPRS Journal of Photogrammetry and Remote Sensing, 2023

Paper Code Data