When Less Is More: Simplicity Beats Complexity for Physics-Constrained InSAR Phase Unwrapping

2026-04-27

AI Summary

This study addresses the challenge of high computational complexity in InSAR phase unwrapping for volcano and earthquake monitoring, where existing models often neglect the smoothness prior inherent in geophysical fields. Through large-scale architectural ablation experiments, the work reveals—for the first time—a “complexity penalty” phenomenon: a lightweight U-Net (7.76M parameters) significantly outperforms various attention-based models on the LiCSAR global benchmark, achieving an R² of 0.834 and an RMSE of 1.01 cm, with an inference latency of only 2.92 ms—satisfying sub-100-ms early-warning requirements. Further power spectral density analysis provides a physical explanation for the effectiveness of simple convolutional architectures under smoothness constraints.

Abstract

Operational phase unwrapping is the primary computational bottleneck in InSAR-based volcanic and seismic monitoring. We challenge the industry trend of adopting high-complexity computer vision architectures, such as attention mechanisms, without validating their suitability for physics-constrained geophysical regression. We present the first large-scale architectural ablation study on a global LiCSAR benchmark (20 frames, 39,724 patches, 651M pixels). Our results reveal a significant "complexity penalty": a vanilla U-Net (7.76M parameters) achieves $R^2=0.834$ and RMSE $= 1.01$ cm, outperforming 11.37M-parameter attention-based models by 34% in $R^2$ and 51% in RMSE. Power Spectral Density (PSD) analysis provides the physical justification: while attention excels at capturing sharp semantic edges in natural images, it injects unphysical high-frequency artifacts ($>0.3$ cycles/pixel) into geophysical fields, violating the fundamental smoothness constraints of elastic surface deformation. With a 2.92ms inference latency (a $2.5\times$ speedup), the vanilla U-Net is the only candidate to comfortably meet the sub-100ms requirement for operational early-warning systems. This work bridges the "publication-to-practice" gap by proving that convolutional locality outperforms modern complexity for smooth-field regression, advocating for physics-informed simplicity in ML4RS. Code available at https://github.com/prabhjotschugh/When-Less-is-More-InSAR-Phase-Unwrapping

Problem

• InSAR phase unwrapping
• physics-constrained regression
• model complexity
• geophysical smoothness
• operational monitoring

Innovation

• physics-constrained learning
• InSAR phase unwrapping
• architectural simplicity
• U-Net
• power spectral density analysis