A Change Detection Reality Check
Not on the map. Why?
Abstract (arXiv)
In recent years, there has been an explosion of proposed change detection deep learning architectures in the remote sensing literature. These approaches claim to offer state-of-the-art performance on different standard benchmark datasets. However, has the field truly made significant progress? In this paper we perform experiments which conclude a simple U-Net segmentation baseline without training tricks or complicated architectural changes is still a top performer for the task of change detection.
Key facts. Cited by 13 papers in the map (data through 7 Oct 2026). Builds on 27 of its 69 references that are in the map. First posted on arXiv 10 Feb 2024.
Builds on
27 of 69 references.
References · 27
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
- 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
- 2022Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language modelscited by 1,970
- 2017Deep Reinforcement Learning that Matterscited by 727
- 2020A Metric Learning Reality Checkcited by 317
- 2018Fully Convolutional Siamese Networks for Change Detectioncited by 305
- 2021Remote Sensing Image Change Detection with Transformerscited by 305
- 2022A Transformer-Based Siamese Network for Change Detectioncited by 243
- 2021Revisiting ResNets: Improved Training and Scaling Strategiescited by 158
- 2019xBD: A Dataset for Assessing Building Damage from Satellite Imagerycited by 138
- 2023GEO-Bench: Toward Foundation Models for Earth Monitoringcited by 122
15 more
- 2019Building Change Detection for Remote Sensing Images Using a Dual Task Constrained Deep Siamese Convolutional Network Modelcited by 84
- 2022Changer: Feature Interaction is What You Need for Change Detectioncited by 78
- 2021S2Looking: A Satellite Side-Looking Dataset for Building Change Detectioncited by 77
- 2021TorchGeo: Deep Learning With Geospatial Datacited by 73
- 2022TINYCD: A (Not So) Deep Learning Model For Change Detectioncited by 35
- 2020Semantic Change Detection with Asymmetric Siamese Networkscited by 31
- 2022Revisiting Consistency Regularization for Semi-supervised Change Detection in Remote Sensing Imagescited by 30
- 2021Unsupervised Domain Adaptation: A Reality Checkcited by 24
- 2021Revisiting 3D ResNets for Video Recognitioncited by 10
- 2024CaBuAr: California Burned Areas dataset for delineationcited by 9
- 2023MS-Former: Memory-Supported Transformer for Weakly Supervised Change Detection with Patch-Level Annotationscited by 6
- 2023Adapting Vision Transformer for Efficient Change Detectioncited by 5
- 2024HiCD: Change Detection in Quality-Varied Images via Hierarchical Correlation Distillationcited by 4
- 2024A simple, strong baseline for building damage detection on the xBD datasetcited by 4
- 2024BD-MSA: Body decouple VHR Remote Sensing Image Change Detection method guided by multi-scale feature information aggregationcited by 1
Built on by
Citing papers per year: 2024: 3; 2025: 7; 2026: 3.
13 papers.
Sort:
- 2024Changen2: Multi-Temporal Remote Sensing Generative Change Foundation Modelcited by 37
- 2025The Change You Want To Detect: Semantic Change Detection In Earth Observation With Hybrid Data Generationcited by 12
- 2025A Survey of Sample-Efficient Deep Learning for Change Detection in Remote Sensing: Tasks, Strategies, and Challengescited by 9
- 2024KAN You See It? KANs and Sentinel for Effective and Explainable Crop Field Segmentationcited by 8
- 2025Be the Change You Want to See: Revisiting Remote Sensing Change Detection Practicescited by 6
- 2024Continuous Urban Change Detection from Satellite Image Time Series with Temporal Feature Refinement and Multi-Task Integrationcited by 5
- 2026PRUE: A Practical Recipe for Field Boundary Segmentation at Scalecited by 5
- 2026Remote Sensing Change Detection via Weak Temporal Supervisioncited by 1
- 2025Deep Self-Supervised Disturbance Mapping with the OPERA Sentinel-1 Radiometric Terrain Corrected SAR Backscatter Productcited by 0
- 2025Operational Change Detection for Geographical Information: Overview and Challengescited by 0
- 2025Leveraging Satellite Image Time Series for Accurate Extreme Event Detectioncited by 0
- 2025Habitat and Land Cover Change Detection in Alpine Protected Areas: A Comparison of AI Architecturescited by 0
- 2026A comprehensive and trustworthy benchmark of AI methods for change detection in Earth observation28 Aug 2026
- 2026PRUE: A Practical Recipe for Field Boundary Segmentation at Scale28 Mar 2026
- 2026Remote Sensing Change Detection via Weak Temporal Supervision5 Jan 2026
- 2025Habitat and Land Cover Change Detection in Alpine Protected Areas: A Comparison of AI Architectures29 Oct 2025
- 2025Be the Change You Want to See: Revisiting Remote Sensing Change Detection Practices4 Jul 2025
- 2025Leveraging Satellite Image Time Series for Accurate Extreme Event Detection13 Jun 2025
- 2025The Change You Want To Detect: Semantic Change Detection In Earth Observation With Hybrid Data Generation19 Mar 2025
- 2025Operational Change Detection for Geographical Information: Overview and Challenges18 Mar 2025
- 2025A Survey of Sample-Efficient Deep Learning for Change Detection in Remote Sensing: Tasks, Strategies, and Challenges5 Feb 2025
- 2025Deep Self-Supervised Disturbance Mapping with the OPERA Sentinel-1 Radiometric Terrain Corrected SAR Backscatter Product15 Jan 2025
- 2024KAN You See It? KANs and Sentinel for Effective and Explainable Crop Field Segmentation13 Aug 2024
- 2024Changen2: Multi-Temporal Remote Sensing Generative Change Foundation Model26 Jun 2024
- 2024Changen2: Multi-Temporal Remote Sensing Generative Change Foundation Modelcited by 37
- 2025The Change You Want To Detect: Semantic Change Detection In Earth Observation With Hybrid Data Generationcited by 12
- 2025A Survey of Sample-Efficient Deep Learning for Change Detection in Remote Sensing: Tasks, Strategies, and Challengescited by 9
- 2024KAN You See It? KANs and Sentinel for Effective and Explainable Crop Field Segmentationcited by 8
- 2025Be the Change You Want to See: Revisiting Remote Sensing Change Detection Practicescited by 6
- 2024Continuous Urban Change Detection from Satellite Image Time Series with Temporal Feature Refinement and Multi-Task Integrationcited by 5
- 2026PRUE: A Practical Recipe for Field Boundary Segmentation at Scalecited by 5
- 2026Remote Sensing Change Detection via Weak Temporal Supervisioncited by 1
- 2025Deep Self-Supervised Disturbance Mapping with the OPERA Sentinel-1 Radiometric Terrain Corrected SAR Backscatter Productcited by 0
- 2025Operational Change Detection for Geographical Information: Overview and Challengescited by 0
- 2025Leveraging Satellite Image Time Series for Accurate Extreme Event Detectioncited by 0
- 2025Habitat and Land Cover Change Detection in Alpine Protected Areas: A Comparison of AI Architecturescited by 0
12 most cited of 13 All 13 (JSON)
- 2026A comprehensive and trustworthy benchmark of AI methods for change detection in Earth observation28 Aug 2026
- 2026PRUE: A Practical Recipe for Field Boundary Segmentation at Scale28 Mar 2026
- 2026Remote Sensing Change Detection via Weak Temporal Supervision5 Jan 2026
- 2025Habitat and Land Cover Change Detection in Alpine Protected Areas: A Comparison of AI Architectures29 Oct 2025
- 2025Be the Change You Want to See: Revisiting Remote Sensing Change Detection Practices4 Jul 2025
- 2025Leveraging Satellite Image Time Series for Accurate Extreme Event Detection13 Jun 2025
- 2025The Change You Want To Detect: Semantic Change Detection In Earth Observation With Hybrid Data Generation19 Mar 2025
- 2025Operational Change Detection for Geographical Information: Overview and Challenges18 Mar 2025
- 2025A Survey of Sample-Efficient Deep Learning for Change Detection in Remote Sensing: Tasks, Strategies, and Challenges5 Feb 2025
- 2025Deep Self-Supervised Disturbance Mapping with the OPERA Sentinel-1 Radiometric Terrain Corrected SAR Backscatter Product15 Jan 2025
- 2024KAN You See It? KANs and Sentinel for Effective and Explainable Crop Field Segmentation13 Aug 2024
- 2024Changen2: Multi-Temporal Remote Sensing Generative Change Foundation Model26 Jun 2024
12 newest of 13 All 13 (JSON)