Continual-MEGA: A Large-scale Benchmark for Generalizable Continual Anomaly Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lee, Geonu, Oh, Yujeong, Jang, Geonhui, Lee, Soyoung, Song, Jeonghyo, Cha, Sungmin, Yoo, YoungJoon
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918325571813376
author Lee, Geonu
Oh, Yujeong
Jang, Geonhui
Lee, Soyoung
Song, Jeonghyo
Cha, Sungmin
Yoo, YoungJoon
author_facet Lee, Geonu
Oh, Yujeong
Jang, Geonhui
Lee, Soyoung
Song, Jeonghyo
Cha, Sungmin
Yoo, YoungJoon
contents In this paper, we introduce a new benchmark for continual learning in anomaly detection, aimed at better reflecting real-world deployment scenarios. Our benchmark, Continual-MEGA, includes a large and diverse dataset that significantly expands existing evaluation settings by combining carefully curated existing datasets with our newly proposed dataset, ContinualAD. In addition to standard continual learning with expanded quantity, we propose a novel scenario that measures zero-shot generalization to unseen classes, those not observed during continual adaptation. This setting poses a new problem setting that continual adaptation also enhances zero-shot performance. We also present a unified baseline algorithm that improves robustness in few-shot detection and maintains strong generalization. Through extensive evaluations, we report three key findings: (1) existing methods show substantial room for improvement, particularly in pixel-level defect localization; (2) our proposed method consistently outperforms prior approaches; and (3) the newly introduced ContinualAD dataset enhances the performance of strong anomaly detection models. We release the benchmark and code in https://github.com/Continual-Mega/Continual-Mega.
format Preprint
id arxiv_https___arxiv_org_abs_2506_00956
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Continual-MEGA: A Large-scale Benchmark for Generalizable Continual Anomaly Detection
Lee, Geonu
Oh, Yujeong
Jang, Geonhui
Lee, Soyoung
Song, Jeonghyo
Cha, Sungmin
Yoo, YoungJoon
Computer Vision and Pattern Recognition
In this paper, we introduce a new benchmark for continual learning in anomaly detection, aimed at better reflecting real-world deployment scenarios. Our benchmark, Continual-MEGA, includes a large and diverse dataset that significantly expands existing evaluation settings by combining carefully curated existing datasets with our newly proposed dataset, ContinualAD. In addition to standard continual learning with expanded quantity, we propose a novel scenario that measures zero-shot generalization to unseen classes, those not observed during continual adaptation. This setting poses a new problem setting that continual adaptation also enhances zero-shot performance. We also present a unified baseline algorithm that improves robustness in few-shot detection and maintains strong generalization. Through extensive evaluations, we report three key findings: (1) existing methods show substantial room for improvement, particularly in pixel-level defect localization; (2) our proposed method consistently outperforms prior approaches; and (3) the newly introduced ContinualAD dataset enhances the performance of strong anomaly detection models. We release the benchmark and code in https://github.com/Continual-Mega/Continual-Mega.
title Continual-MEGA: A Large-scale Benchmark for Generalizable Continual Anomaly Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.00956