A One-Shot Texture-Perceiving Generative Adversarial Network for Unsupervised Surface Inspection

TL;DR

This paper introduces HTP-GAN, a hierarchical one-shot unsupervised model for surface defect detection that captures both global and local textures to identify anomalies with minimal training data.

Contribution

The paper proposes a novel hierarchical texture-perceiving GAN architecture that learns from a single normal image for unsupervised surface inspection, reducing reliance on large datasets.

Findings

Effective defect detection across multiple datasets

Outperforms existing unsupervised methods

Capable of identifying subtle surface defects

Abstract

Visual surface inspection is a challenging task owing to the highly diverse appearance of target surfaces and defective regions. Previous attempts heavily rely on vast quantities of training examples with manual annotation. However, in some practical cases, it is difficult to obtain a large number of samples for inspection. To combat it, we propose a hierarchical texture-perceiving generative adversarial network (HTP-GAN) that is learned from the one-shot normal image in an unsupervised scheme. Specifically, the HTP-GAN contains a pyramid of convolutional GANs that can capture the global structure and fine-grained representation of an image simultaneously. This innovation helps distinguishing defective surface regions from normal ones. In addition, in the discriminator, a texture-perceiving module is devised to capture the spatially invariant representation of normal image via directional convolutions, making it more sensitive to defective areas. Experiments on a variety of datasets consistently demonstrate the effectiveness of our method.