SuperAD: A Training-free Anomaly Classification and Segmentation Method for CVPR 2025 VAND 3.0 Workshop Challenge Track 1: Adapt & Detect

TL;DR

SuperAD is a training-free anomaly detection and segmentation method leveraging DINOv2 features, designed to handle complex real-world industrial anomalies with high accuracy without requiring model training.

Contribution

The paper introduces SuperAD, a novel training-free approach that uses feature extraction and nearest neighbor matching for anomaly detection and segmentation in industrial settings.

Findings

Achieves competitive results on MVTec AD 2 dataset

Handles complex anomalies like transparency and occlusions

Operates without training, reducing computational costs

Abstract

In this technical report, we present our solution to the CVPR 2025 Visual Anomaly and Novelty Detection (VAND) 3.0 Workshop Challenge Track 1: Adapt & Detect: Robust Anomaly Detection in Real-World Applications. In real-world industrial anomaly detection, it is crucial to accurately identify anomalies with physical complexity, such as transparent or reflective surfaces, occlusions, and low-contrast contaminations. The recently proposed MVTec AD 2 dataset significantly narrows the gap between publicly available benchmarks and anomalies found in real-world industrial environments. To address the challenges posed by this dataset--such as complex and varying lighting conditions and real anomalies with large scale differences--we propose a fully training-free anomaly detection and segmentation method based on feature extraction using the DINOv2 model named SuperAD. Our method carefully selects a small number of normal reference images and constructs a memory bank by leveraging the strong representational power of DINOv2. Anomalies are then segmented by performing nearest neighbor matching between test image features and the memory bank. Our method achieves competitive results on both test sets of the MVTec AD 2 dataset.