Diachronic Stereo Matching for Multi-Date Satellite Imagery

TL;DR

This work tackles reliable 3D reconstruction from satellite image pairs captured at distant dates, where appearance changes break conventional stereo assumptions. It shows that fine-tuning a monocular-prior stereo network (MonSter) on a carefully curated diachronic dataset, together with RPC-based rectification, enables robust disparity estimation and DSM reconstruction across time. The authors release a diachronic–synchronic training set derived from DFC2019 Track 3 and demonstrate state-of-the-art MAE performance across multiple test sites, reducing errors on diachronic pairs while preserving synchronic accuracy. The approach offers practical value for time-series geospatial analysis and change monitoring, though it remains affected by vegetation biases and temporal discordances between LiDAR ground truth and satellite imagery.

Abstract

Recent advances in image-based satellite 3D reconstruction have progressed along two complementary directions. On one hand, multi-date approaches using NeRF or Gaussian-splatting jointly model appearance and geometry across many acquisitions, achieving accurate reconstructions on opportunistic imagery with numerous observations. On the other hand, classical stereoscopic reconstruction pipelines deliver robust and scalable results for simultaneous or quasi-simultaneous image pairs. However, when the two images are captured months apart, strong seasonal, illumination, and shadow changes violate standard stereoscopic assumptions, causing existing pipelines to fail. This work presents the first Diachronic Stereo Matching method for satellite imagery, enabling reliable 3D reconstruction from temporally distant pairs. Two advances make this possible: (1) fine-tuning a state-of-the-art deep stereo network that leverages monocular depth priors, and (2) exposing it to a dataset specifically curated to include a diverse set of diachronic image pairs. In particular, we start from a pretrained MonSter model, trained initially on a mix of synthetic and real datasets such as SceneFlow and KITTI, and fine-tune it on a set of stereo pairs derived from the DFC2019 remote sensing challenge. This dataset contains both synchronic and diachronic pairs under diverse seasonal and illumination conditions. Experiments on multi-date WorldView-3 imagery demonstrate that our approach consistently surpasses classical pipelines and unadapted deep stereo models on both synchronic and diachronic settings. Fine-tuning on temporally diverse images, together with monocular priors, proves essential for enabling 3D reconstruction from previously incompatible acquisition dates. Left image (winter) Right image (autumn) DSM geometry Ours (1.23 m) Zero-shot (3.99 m) LiDAR GT Figure 1. Output geometry for a winter-autumn image pair from Omaha (OMA 331 test scene). Our method recovers accurate geometry despite the diachronic nature of the pair, exhibiting strong appearance changes, which cause existing zero-shot methods to fail. Missing values due to perspective shown in black. Mean altitude error in parentheses; lower is better.

Figures (8)

• Figure 1: Output geometry for a winter-autumn image pair from Omaha (OMA_331 test scene). Our method recovers accurate geometry despite the diachronic nature of the pair, exhibiting strong appearance changes, which cause existing zero-shot methods to fail. Missing values due to perspective shown in black. Mean altitude error in parentheses; lower is better.
• Figure 2: Comparison of SIFT and DISK+LightGlue matches for the diachronic pair OMA_331_017 -- OMA_331_036.
• Figure 3: Monocular depth consistency across seasons. Despite large radiometric differences between seasonal images from the Omaha OMA_893 scene, the monocular Depth Anything V2 predictions exhibit coherent geometric structure.
• Figure 4: Qualitative results of disparity predictions on a selection of hard diachronic image pairs from the test set, listed in Table \ref{['tab:grid_oma']}.
• Figure 5: Top to bottom: DSMs produced by the classical s2p-hd pipeline, zero-shot MonSter (mix all), our method, and the ground-truth DSM on a selection of hard diachronic image pairs from the test set, listed in Table \ref{['tab:grid_oma']}. Missing values shown in black.