SinSEMI: A One-Shot Image Generation Model and Data-Efficient Evaluation Framework for Semiconductor Inspection Equipment

TL;DR

SinSEMI is a novel one-shot image generation model that creates realistic semiconductor inspection images from a single input, supported by a new evaluation framework suitable for data-scarce scenarios.

Contribution

The paper introduces SinSEMI, a flow-based one-shot image generation model with perceptual guidance and a specialized evaluation framework for semiconductor inspection images.

Findings

SinSEMI outperforms existing methods in visual quality and diversity.

Generated images are effective for training semiconductor AI models.

The evaluation framework requires only two reference images for assessment.

Abstract

In the early stages of semiconductor equipment development, obtaining large quantities of raw optical images poses a significant challenge. This data scarcity hinder the advancement of AI-powered solutions in semiconductor manufacturing. To address this challenge, we introduce SinSEMI, a novel one-shot learning approach that generates diverse and highly realistic images from single optical image. SinSEMI employs a multi-scale flow-based model enhanced with LPIPS (Learned Perceptual Image Patch Similarity) energy guidance during sampling, ensuring both perceptual realism and output variety. We also introduce a comprehensive evaluation framework tailored for this application, which enables a thorough assessment using just two reference images. Through the evaluation against multiple one-shot generation techniques, we demonstrate SinSEMI's superior performance in visual quality, quantitative measures, and downstream tasks. Our experimental results demonstrate that SinSEMI-generated images achieve both high fidelity and meaningful diversity, making them suitable as training data for semiconductor AI applications.