MicroFlow: Domain-Specific Optical Flow for Ground Deformation Estimation in Seismic Events

2025-04-18

AI Summary

Addressing the challenge of sub-pixel surface deformation estimation from optical satellite imagery in earthquake scenarios—particularly under conditions of ground-truth label scarcity, low sensitivity to small displacements, and strong spatiotemporal interference from geological and anthropogenic factors—this paper proposes an iterative explicit deformation correction framework. Methodologically, it integrates three key innovations: (1) a correlation-agnostic backbone network that decouples feature extraction from deformation modeling; (2) a non-convex total variation regularizer that preserves sharp fault edges while ensuring regional smoothness; and (3) an explicit warping layer coupled with a self-supervised training strategy for end-to-end displacement field optimization. Evaluated on a semi-synthetic benchmark, the method significantly outperforms conventional geophysical approaches. Moreover, it demonstrates strong generalization capability and sub-pixel accuracy on medium-to-high-resolution real-world remote sensing imagery.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Intelligent Robots: State Estimation
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments

Abstract

Dense ground displacement measurements are crucial for geological studies but are impractical to collect directly. Traditionally, displacement fields are estimated using patch matching on optical satellite images from different acquisition times. While deep learning-based optical flow models are promising, their adoption in ground deformation analysis is hindered by challenges such as the absence of real ground truth, the need for sub-pixel precision, and temporal variations due to geological or anthropogenic changes. In particular, we identify that deep learning models relying on explicit correlation layers struggle at estimating small displacements in real-world conditions. Instead, we propose a model that employs iterative refinements with explicit warping layers and a correlation-independent backbone, enabling sub-pixel precision. Additionally, a non-convex variant of Total Variation regularization preserves fault-line sharpness while maintaining smoothness elsewhere. Our model significantly outperforms widely used geophysics methods on semi-synthetic benchmarks and generalizes well to challenging real-world scenarios captured by both medium- and high-resolution sensors. Project page: https://jbertrand89.github.io/microflow/.

Problem

• Estimating ground deformation in seismic events accurately
• Overcoming sub-pixel precision challenges in optical flow
• Preserving fault-line sharpness with non-convex regularization

Innovation

• Iterative refinements with explicit warping layers
• Correlation-independent backbone for sub-pixel precision
• Non-convex Total Variation regularization preserves fault-lines