Wavefront-Constrained Passive Obscured Object Detection

TL;DR

This paper introduces WavePCNet, a physics-driven neural network that models wavefront propagation to improve detection of obscured objects in challenging conditions, outperforming existing methods in accuracy and robustness.

Contribution

The paper presents a novel WavePCNet that integrates complex wavefront modeling, a momentum memory mechanism, and frequency-selective enhancement for better obscured object detection.

Findings

WavePCNet outperforms state-of-the-art methods in accuracy.

WavePCNet demonstrates superior robustness in complex environments.

The approach effectively models coherent light propagation physics.

Abstract

Accurately localizing and segmenting obscured objects from faint light patterns beyond the field of view is highly challenging due to multiple scattering and medium-induced perturbations. Most existing methods, based on real-valued modeling or local convolutional operations, are inadequate for capturing the underlying physics of coherent light propagation. Moreover, under low signal-to-noise conditions, these methods often converge to non-physical solutions, severely compromising the stability and reliability of the observation. To address these challenges, we propose a novel physics-driven Wavefront Propagating Compensation Network (WavePCNet) to simulate wavefront propagation and enhance the perception of obscured objects. This WavePCNet integrates the Tri-Phase Wavefront Complex-Propagation Reprojection (TriWCP) to incorporate complex amplitude transfer operators to precisely constrain coherent propagation behavior, along with a momentum memory mechanism to effectively suppress the accumulation of perturbations. Additionally, a High-frequency Cross-layer Compensation Enhancement is introduced to construct frequency-selective pathways with multi-scale receptive fields and dynamically model structural consistency across layers, further boosting the model's robustness and interpretability under complex environmental conditions. Extensive experiments conducted on four physically collected datasets demonstrate that WavePCNet consistently outperforms state-of-the-art methods across both accuracy and robustness.