Overview
Field: Computer Vision Authors: Gijung Lee, Ronald Wilson, Damon L. Woodard Published: 2026-08-12 arXiv: 2508.05150
Key points
- Hardware assurance depends on SEM imaging of nanoscale structures, but assembling large datasets for automated analysis is blocked by costly acquisition and strict IP constraints on proprietary designs.
- The proposed privacy-preserving pipeline distorts the functional design heavily to secure IP, while synthesizing a visually realistic dataset from a small set of initial examples.
- A StyleGAN learns the distribution of hardware layout masks to generate novel, macroscopically varied structures.
- A conditional GAN (Pix2PixHD) translates these masks into realistic SEM images that preserve authentic textures and noise.
- Main finding: a segmentation model trained only on this synthetic data achieves sim-to-real transfer to real images and outperforms baselines trained on limited real datasets.
- Since the synthetic layouts are novel and do not replicate any specific proprietary routing, deploying the model reduces the risk of sensitive IP exposure to attacks such as gradient inversion and membership inference.
Original abstract (excerpt)
> Hardware assurance relies on scanning electron microscopy (SEM) to verify nanoscale structures, but assembling the large, high-quality datasets required for automated analysis is impeded by time-intensive acquisition and strict intellectual property (IP) constraints on proprietary designs. We propose a privacy-preserving pipeline that secures IP by heavily distorting the functional design while generating a visually realistic synthetic dataset from a small set of initial examples. A StyleGAN first learns the distribution of hardware layout masks to generate novel, macroscopically varied structures. Subsequently, a conditional GAN (Pix2PixHD) translates these masks into realistic SEM images that preserve authentic textures and noise. The primary finding of this work is that a segmentation m...
---
*Auto-collected on 2026-08-12*