Few Shot Part Segmentation Reveals Compositional Logic for Industrial Anomaly Detection

TL;DR

This paper introduces a few-shot component segmentation model that leverages limited labeled data and unlabeled images to improve industrial anomaly detection by reasoning about component composition and logical constraints.

Contribution

The study presents a novel segmentation approach using histogram matching, entropy loss, and memory banks to enhance anomaly detection in industrial images with minimal supervision.

Findings

Achieves 98.1% AUROC on MVTec LOCO AD benchmark.

Outperforms existing methods with 89.6% AUROC.

Effectively detects logical anomalies with few labeled samples.

Abstract

Logical anomalies (LA) refer to data violating underlying logical constraints e.g., the quantity, arrangement, or composition of components within an image. Detecting accurately such anomalies requires models to reason about various component types through segmentation. However, curation of pixel-level annotations for semantic segmentation is both time-consuming and expensive. Although there are some prior few-shot or unsupervised co-part segmentation algorithms, they often fail on images with industrial object. These images have components with similar textures and shapes, and a precise differentiation proves challenging. In this study, we introduce a novel component segmentation model for LA detection that leverages a few labeled samples and unlabeled images sharing logical constraints. To ensure consistent segmentation across unlabeled images, we employ a histogram matching loss in conjunction with an entropy loss. As segmentation predictions play a crucial role, we propose to enhance both local and global sample validity detection by capturing key aspects from visual semantics via three memory banks: class histograms, component composition embeddings and patch-level representations. For effective LA detection, we propose an adaptive scaling strategy to standardize anomaly scores from different memory banks in inference. Extensive experiments on the public benchmark MVTec LOCO AD reveal our method achieves 98.1% AUROC in LA detection vs. 89.6% from competing methods.