# A Comparative Study on Deep-Learning Methods for Dense Image Matching of Multi-angle and Multi-date Remote Sensing Stereo Images

arXiv: https://arxiv.org/abs/2210.14031 · HTML: https://mlsanity.com/paper/2210.14031/

## Key facts

- Cited by 5 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 25 Oct 2022.

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

- [Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition](https://mlsanity.com/paper/1406.4729/) (2014)
- [Pyramid Stereo Matching Network](https://mlsanity.com/paper/1803.08669/) (2018)
- [Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matching](https://mlsanity.com/paper/1912.06378/) (2019)
- [End-to-End Learning of Geometry and Context for Deep Stereo Regression](https://mlsanity.com/paper/1703.04309/) (2017)
- [GA-Net: Guided Aggregation Net for End-to-end Stereo Matching](https://mlsanity.com/paper/1904.06587/) (2019)
- [AANet: Adaptive Aggregation Network for Efficient Stereo Matching](https://mlsanity.com/paper/2004.09548/) (2020)
- [Hierarchical Neural Architecture Search for Deep Stereo Matching](https://mlsanity.com/paper/2010.13501/) (2020)
- [Stereo Matching by Training a Convolutional Neural Network to Compare Image Patches](https://mlsanity.com/paper/1510.05970/) (2015)
- [Domain-invariant Stereo Matching Networks](https://mlsanity.com/paper/1911.13287/) (2019)
- [Computing the Stereo Matching Cost with a Convolutional Neural Network](https://mlsanity.com/paper/1409.4326/) (2014)
- [Semantic Stereo for Incidental Satellite Images](https://mlsanity.com/paper/1811.08739/) (2018)
- [AdaStereo: A Simple and Efficient Approach for Adaptive Stereo Matching](https://mlsanity.com/paper/2004.04627/) (2020)
- [Zoom and Learn: Generalizing Deep Stereo Matching to Novel Domains](https://mlsanity.com/paper/1803.06641/) (2018)
- [A New Stereo Benchmarking Dataset for Satellite Images](https://mlsanity.com/paper/1907.04404/) (2019)
- [A Survey on Deep Learning Architectures for Image-based Depth Reconstruction](https://mlsanity.com/paper/1906.06113/) (2019)
- [Automated 3D recovery from very high resolution multi-view satellite images](https://mlsanity.com/paper/1905.07475/) (2019)
- [3D Iterative Spatiotemporal Filtering for Classification of Multitemporal Satellite Data Sets](https://mlsanity.com/paper/2107.00590/) (2021)
- [A Comparison of Stereo-Matching Cost between Convolutional Neural Network and Census for Satellite Images](https://mlsanity.com/paper/1905.09147/) (2019)

## Built on by (most cited first, 5 of 5)

- [Multi-tiling Neural Radiance Field (NeRF) -- Geometric Assessment on Large-scale Aerial Datasets](https://mlsanity.com/paper/2310.00530/) (2023)
- [An Evaluation of DUSt3R/MASt3R/VGGT 3D Reconstruction on Photogrammetric Aerial Blocks](https://mlsanity.com/paper/2507.14798/) (2025)
- [An evaluation of Deep Learning based stereo dense matching dataset shift from aerial images and a large scale stereo dataset](https://mlsanity.com/paper/2402.12522/) (2024)
- [Deep Learning Meets Satellite Images -- An Evaluation on Handcrafted and Learning-based Features for Multi-date Satellite Stereo Images](https://mlsanity.com/paper/2409.02825/) (2024)
- [SeasonStereo: Robust Dense Stereo Matching for Multi-Date Satellite Imagery via Generative AI](https://mlsanity.com/paper/2607.27139/) (2026)

## Abstract (arXiv)

Deep learning (DL) stereo matching methods gained great attention in remote sensing satellite datasets. However, most of these existing studies conclude assessments based only on a few/single stereo images lacking a systematic evaluation on how robust DL methods are on satellite stereo images with varying radiometric and geometric configurations. This paper provides an evaluation of four DL stereo matching methods through hundreds of multi-date multi-site satellite stereo pairs with varying geometric configurations, against the traditional well-practiced Census-SGM (Semi-global matching), to comprehensively understand their accuracy, robustness, generalization capabilities, and their practical potential. The DL methods include a learning-based cost metric through convolutional neural networks (MC-CNN) followed by SGM, and three end-to-end (E2E) learning models using Geometry and Context Network (GCNet), Pyramid Stereo Matching Network (PSMNet), and LEAStereo. Our experiments show that E2E algorithms can achieve upper limits of geometric accuracies, while may not generalize well for unseen data. The learning-based cost metric and Census-SGM are rather robust and can consistently achieve acceptable results. All DL algorithms are robust to geometric configurations of stereo pairs and are less sensitive in comparison to the Census-SGM, while learning-based cost metrics can generalize on satellite images when trained on different datasets (airborne or ground-view).