Feature Perturbation Pool-based Fusion Network for Unified Multi-Class Industrial Defect Detection

TL;DR

This paper introduces FPFNet, a unified multi-class industrial defect detection network that enhances robustness and accuracy by integrating a stochastic feature perturbation pool with multi-layer feature fusion, without increasing model complexity.

Contribution

The proposed FPFNet combines a feature perturbation pool with multi-layer feature fusion to improve multi-class defect detection robustness and accuracy in a unified framework.

Findings

Achieves 97.17% image-level AUROC on MVTec-AD

Attains 96.93% pixel-level AUROC on MVTec-AD

Surpasses existing methods with no additional parameters or complexity.

Abstract

Multi-class defect detection constitutes a critical yet challenging task in industrial quality inspection, where existing approaches typically suffer from two fundamental limitations: (i) the necessity of training separate models for each defect category, resulting in substantial computational and memory overhead, and (ii) degraded robustness caused by inter-class feature perturbation when heterogeneous defect categories are jointly modeled. In this paper, we present FPFNet, a Feature Perturbation Pool-based Fusion Network that synergistically integrates a stochastic feature perturbation pool with a multi-layer feature fusion strategy to address these challenges within a unified detection framework. The feature perturbation pool enriches the training distribution by randomly injecting diverse noise patterns -- including Gaussian noise, F-Noise, and F-Drop -- into the extracted feature representations, thereby strengthening the model's robustness against domain shifts and unseen defect morphologies. Concurrently, the multi-layer feature fusion module aggregates hierarchical feature representations from both the encoder and decoder through residual connections and normalization, enabling the network to capture complex cross-scale relationships while preserving fine-grained spatial details essential for precise defect localization. Built upon the UniAD architecture~\cite{you2022unified}, our method achieves state-of-the-art performance on two widely adopted benchmarks: 97.17\% image-level AUROC and 96.93\% pixel-level AUROC on MVTec-AD, and 91.08\% image-level AUROC and 99.08\% pixel-level AUROC on VisA, surpassing existing methods by notable margins while introducing no additional learnable parameters or computational complexity.