# Physically-Guided Optical Inversion Enable Non-Contact Side-Channel Attack on Isolated Screens

arXiv: https://arxiv.org/abs/2604.13419 · HTML: https://mlsanity.com/paper/2604.13419/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 43 references that are in the map.
- First posted on arXiv 15 Apr 2026.

## Builds on (most cited first, 21 of 21)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Image-to-Image Translation with Conditional Adversarial Networks](https://mlsanity.com/paper/1611.07004/) (2016)
- [Uformer: A General U-Shaped Transformer for Image Restoration](https://mlsanity.com/paper/2106.03106/) (2021)
- [Toward Multimodal Image-to-Image Translation](https://mlsanity.com/paper/1711.11586/) (2017)
- [Curricular Contrastive Regularization for Physics-aware Single Image Dehazing](https://mlsanity.com/paper/2303.14218/) (2023)
- [HVI: A New Color Space for Low-light Image Enhancement](https://mlsanity.com/paper/2502.20272/) (2025)
- [DarkIR: Robust Low-Light Image Restoration](https://mlsanity.com/paper/2412.13443/) (2024)
- [GCFSR: a Generative and Controllable Face Super Resolution Method Without Facial and GAN Priors](https://mlsanity.com/paper/2203.07319/) (2022)
- [DivCo: Diverse Conditional Image Synthesis via Contrastive Generative Adversarial Network](https://mlsanity.com/paper/2103.07893/) (2021)
- [Styleformer: Transformer based Generative Adversarial Networks with Style Vector](https://mlsanity.com/paper/2106.07023/) (2021)
- [Arbitrary-Scale Image Generation and Upsampling using Latent Diffusion Model and Implicit Neural Decoder](https://mlsanity.com/paper/2403.10255/) (2024)
- [Learning Hazing to Dehazing: Towards Realistic Haze Generation for Real-World Image Dehazing](https://mlsanity.com/paper/2503.19262/) (2025)
- [BEVDiffuser: Plug-and-Play Diffusion Model for BEV Denoising with Ground-Truth Guidance](https://mlsanity.com/paper/2502.19694/) (2025)
- [Iterative Predictor-Critic Code Decoding for Real-World Image Dehazing](https://mlsanity.com/paper/2503.13147/) (2025)
- [PBR-NeRF: Inverse Rendering with Physics-Based Neural Fields](https://mlsanity.com/paper/2412.09680/) (2024)
- [DnLUT: Ultra-Efficient Color Image Denoising via Channel-Aware Lookup Tables](https://mlsanity.com/paper/2503.15931/) (2025)
- [Channel Consistency Prior and Self-Reconstruction Strategy Based Unsupervised Image Deraining](https://mlsanity.com/paper/2503.18703/) (2025)
- [AdaBM: On-the-Fly Adaptive Bit Mapping for Image Super-Resolution](https://mlsanity.com/paper/2404.03296/) (2024)
- [Tokenize Image Patches: Global Context Fusion for Effective Haze Removal in Large Images](https://mlsanity.com/paper/2504.09621/) (2025)
- [CoA: Towards Real Image Dehazing via Compression-and-Adaptation](https://mlsanity.com/paper/2504.05590/) (2025)
- [Deep-TEMPEST: Using Deep Learning to Eavesdrop on HDMI from its Unintended Electromagnetic Emanations](https://mlsanity.com/paper/2407.09717/) (2024)

## Abstract (arXiv)

Noncontact exfiltration of electronic screen content poses a security challenge, with side-channel incursions as the principal vector. We introduce an optical projection side-channel paradigm that confronts two core instabilities: (i) the near-singular Jacobian spectrum of projection mapping breaches Hadamard stability, rendering inversion hypersensitive to perturbations; (ii) irreversible compression in light transport obliterates global semantic cues, magnifying reconstruction ambiguity. Exploiting passive speckle patterns formed by diffuse reflection, our Irradiance Robust Radiometric Inversion Network (IR4Net) fuses a Physically Regularized Irradiance Approximation (PRIrr-Approximation), which embeds the radiative transfer equation in a learnable optimizer, with a contour-to-detail cross-scale reconstruction mechanism that arrests noise propagation. Moreover, an Irreversibility Constrained Semantic Reprojection (ICSR) module reinstates lost global structure through context-driven semantic mapping. Evaluated across four scene categories, IR4Net achieves fidelity beyond competing neural approaches while retaining resilience to illumination perturbations.