Wavefront-Constrained Passive Obscured Object Detection

2025-11-25

AI Summary

Localizing and segmenting occluded objects from weak optical signals outside the field of view remains challenging due to multiple scattering, medium-induced perturbations, and extremely low signal-to-noise ratios. Conventional real-valued modeling or local convolutional approaches fail to capture the fundamental physics of coherent light propagation, often converging to non-physical solutions. Method: We propose WavePCNet, a wavefront propagation compensation network featuring a novel tri-phase complex propagation re-projection module that explicitly models complex-amplitude transmission; a momentum-based memory mechanism to suppress cumulative perturbations; and hybrid high-frequency cross-layer compensation with frequency-selective pathways to enhance structural consistency. Results: Evaluated on four real-world acquired datasets, WavePCNet achieves significant improvements in localization and segmentation accuracy and robustness, consistently outperforming state-of-the-art methods across all benchmarks.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Machine Learning: Multi-instance/Multi-view Learning
• Search and Optimization: Learning to Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs

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.

Problem

• Detecting obscured objects from faint scattered light patterns
• Addressing non-physical solutions under low signal-to-noise conditions
• Modeling coherent light propagation through complex media perturbations

Innovation

• WavePCNet simulates wavefront propagation for obscured objects
• TriWCP integrates complex amplitude transfer operators
• High-frequency compensation enhances robustness and interpretability