Toward Multi-class Anomaly Detection: Exploring Class-aware Unified Model against Inter-class Interference

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Jiang, Xi, Chen, Ying, Nie, Qiang, Liu, Jianlin, Liu, Yong, Wang, Chengjie, Zheng, Feng
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913276369043456
author Jiang, Xi
Chen, Ying
Nie, Qiang
Liu, Jianlin
Liu, Yong
Wang, Chengjie
Zheng, Feng
author_facet Jiang, Xi
Chen, Ying
Nie, Qiang
Liu, Jianlin
Liu, Yong
Wang, Chengjie
Zheng, Feng
contents In the context of high usability in single-class anomaly detection models, recent academic research has become concerned about the more complex multi-class anomaly detection. Although several papers have designed unified models for this task, they often overlook the utility of class labels, a potent tool for mitigating inter-class interference. To address this issue, we introduce a Multi-class Implicit Neural representation Transformer for unified Anomaly Detection (MINT-AD), which leverages the fine-grained category information in the training stage. By learning the multi-class distributions, the model generates class-aware query embeddings for the transformer decoder, mitigating inter-class interference within the reconstruction model. Utilizing such an implicit neural representation network, MINT-AD can project category and position information into a feature embedding space, further supervised by classification and prior probability loss functions. Experimental results on multiple datasets demonstrate that MINT-AD outperforms existing unified training models.
format Preprint
id arxiv_https___arxiv_org_abs_2403_14213
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Toward Multi-class Anomaly Detection: Exploring Class-aware Unified Model against Inter-class Interference
Jiang, Xi
Chen, Ying
Nie, Qiang
Liu, Jianlin
Liu, Yong
Wang, Chengjie
Zheng, Feng
Computer Vision and Pattern Recognition
In the context of high usability in single-class anomaly detection models, recent academic research has become concerned about the more complex multi-class anomaly detection. Although several papers have designed unified models for this task, they often overlook the utility of class labels, a potent tool for mitigating inter-class interference. To address this issue, we introduce a Multi-class Implicit Neural representation Transformer for unified Anomaly Detection (MINT-AD), which leverages the fine-grained category information in the training stage. By learning the multi-class distributions, the model generates class-aware query embeddings for the transformer decoder, mitigating inter-class interference within the reconstruction model. Utilizing such an implicit neural representation network, MINT-AD can project category and position information into a feature embedding space, further supervised by classification and prior probability loss functions. Experimental results on multiple datasets demonstrate that MINT-AD outperforms existing unified training models.
title Toward Multi-class Anomaly Detection: Exploring Class-aware Unified Model against Inter-class Interference
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2403.14213