# PISE: Physics-Anchored Semantically-Enhanced Deep Computational Ghost Imaging for Robust Low-Bandwidth Machine Perception Source: [PISE: Physics-Anchored Semantically-Enhanced Deep Computational Ghost Imaging for Robust Low-Bandwidth Machine Perception]() Collection index: [llms.txt]() Data updated: 2026-03-17T05:00:04.157Z Authors: Tong Wu Published: 2026-01-18T19:31:24.000Z ## 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. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Image Reconstruction on CIFAR-10 5% sampling, Poisson Noise, 8-bit Quant. (test)](): Image Reconstruction; CIFAR-10 5% sampling, Poisson Noise, 8-bit Quant. (test); PSNR (dB): 18.77. - [Image Reconstruction on 28 x 28 Images (test)](): Image Reconstruction; 28 x 28 Images (test); FPS (Hz): 2455. - [Image Classification on CIFAR-10 5% sampling, Poisson Noise, 8-bit Quant. (test)](): Image Classification; CIFAR-10 5% sampling, Poisson Noise, 8-bit Quant. (test); Accuracy: 21.64. ## Original sources and implementation [Original paper]() [PDF]()