PISE: Physics-Anchored Semantically-Enhanced Deep Computational Ghost Imaging for Robust Low-Bandwidth Machine Perception

TL;DR

PISE is a physics-informed deep ghost imaging framework that enhances low-bandwidth machine perception by improving classification accuracy and reducing variance through semantic guidance and adjoint operator initialization.

Contribution

It introduces a novel combination of physics-based initialization and semantic guidance to improve ghost imaging performance in low-bandwidth scenarios.

Findings

Increases classification accuracy by 2.57%.

Reduces variance by 9 times at 5% sampling.

Effective for low-bandwidth edge perception.

Abstract

We propose PISE, a physics-informed deep ghost imaging framework for low-bandwidth edge perception. By combining adjoint operator initialization with semantic guidance, PISE improves classification accuracy by 2.57% and reduces variance by 9x at 5% sampling.