Patch-aware Vector Quantized Codebook Learning for Unsupervised Visual Defect Detection

2025-01-15 · Qisen Cheng, Shuhui Qu, Janghwan Lee

Unsupervised visual defect detection is critical in industrial applications, requiring a representation space that captures normal data features while detecting deviations. Achieving a balance between expressiveness and compactness is challenging; an overly expressive space risks inefficiency and mode collapse, impairing detection accuracy. We propose a novel approach using an enhanced VQ-VAE framework optimized for unsupervised defect detection. Our model introduces a patch-aware dynamic code assignment scheme, enabling context-sensitive code allocation to optimize spatial representation. This strategy enhances normal-defect distinction and improves detection accuracy during inference. Experiments on MVTecAD, BTAD, and MTSD datasets show our method achieves state-of-the-art performance.

📄 PDF Abstract BibTeX arXiv:2501.09187

Code (0)

등록된 구현이 없습니다.

Tasks

Defect Detection

Methods 이 논문이 사용한 방법론

VQ-VAE VQ-VAE is a type of variational autoencoder that uses vector quantisation to obtain a discrete latent representation. It differs from…

Learning Product Codebooks using Vector Quantized Autoencoders for Image Retrieval

2018-07-12 · Hanwei Wu, Markus Flierl

Vector-Quantized Variational Autoencoders (VQ-VAE)[1] provide an unsupervised model for learning discrete representations by combining vector quantization and autoencoders. In this paper, we study the use of VQ-VAE for r…

Image Retrieval Quantization Representation Learning Retrieval

Robust Training of Vector Quantized Bottleneck Models

2020-05-18 · Adrian Łańcucki, Jan Chorowski, Guillaume Sanchez, Ricard Marxer 외

In this paper we demonstrate methods for reliable and efficient training of discrete representation using Vector-Quantized Variational Auto-Encoder models (VQ-VAEs). Discrete latent variable models have been shown to lea…

Clustering Disentanglement Representation Learning Voice Conversion

BeeVe: Unsupervised Acoustic State Discovery in Honey Bee Buzzing

2026-05-08 · Hamze Hammami, Nidhal Abdulaziz arxiv

Discovering structure in biological signals without supervision is a fundamental problem in computational intelligence, yet existing bioacoustic methods assume vocal production models or predefined semantic units, leavin…

COMET: Codebook-based Online-adaptive Multi-scale Embedding for Time-series Anomaly Detection

2026-02-02 · Jinwoo Park, Hyeongwon Kang, Seung Hun Han, Pilsung Kang arxiv

Time series anomaly detection is a critical task across various industrial domains. However, capturing temporal dependencies and multivariate correlations within patch-level representation learning remains underexplored,…

Time Series Anomaly Detection Representation Learning Contrastive Learning

Vector Quantization for Deep-Learning-Based CSI Feedback in Massive MIMO Systems

2024-03-12 · Junyong Shin, Yujin Kang, Yo-Seb Jeon

This paper presents a finite-rate deep-learning (DL)-based channel state information (CSI) feedback method for massive multiple-input multiple-output (MIMO) systems. The presented method provides a finite-bit representat…

Quantization