computer-vision-2

continual anomaly detection

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in continual-anomaly-detection-2

Trend

Dataset

Best Model

Actions

No benchmarks available.

Libraries

i

Use these libraries to find continual-anomaly-detection-2 models and implementations

Datasets

No datasets available.

Subtasks

No subtasks available.

Most implemented papers

ARCADe: A Rapid Continual Anomaly Detector

Volker Tresp, A. Frikha, Denis Krompass•Sun Aug 09 2020

This work proposes A Rapid Continual Anomaly Detector (ARCADe), an approach to train neural networks to be robust against the major challenges of this new learning problem, namely catastrophic forgetting and overfitting to the majority class.

12

Content

0

Paper Graph

Continual Learning for Unsupervised Anomaly Detection in Continuous Auditing of Financial Accounting Data

Marco Schreyer, Damian Borth, Hamed Hemati•Fri Dec 24 2021

This work proposes a continual anomaly detection framework to overcome both challenges and designed to learn from a stream of journal entry data experiences, and provides initial evidence that such a learning scheme offers the ability to reduce false-positive alerts and false-negative decisions.

13 0

Paper Graph

Towards Continual Adaptation in Industrial Anomaly Detection

Jun Liu, Feng Zheng, Chengjie Wang, Bin-Bin Gao, Wu-Jun Li, Jiawei Zhan, Jinbao Wang, Bizhong Xia•Sun Oct 09 2022

A novel method named Distribution of Normal Embeddings (DNE), which utilizes the feature distribution of normal training samples from past tasks and not only effectively alleviates catastrophic forgetting in CAD but also can be integrated with SCL methods to further improve their performance.

32 0

Paper Graph

An Incremental Unified Framework for Small Defect Inspection

Jiaqi Tang, Ruizheng Wu, Xiaogang Xu, Sixing Hu, Ying-Cong Chen, Hao Lu, Tong Zhang, Tsz Wa Cheng, Ming Ge, F. Tsung•Wed Dec 13 2023

The Incremental Unified Framework (IUF) is presented, which can reduce the feature conflict problem when continuously integrating new objects in the pipeline, making it advantageous in object-incremental learning scenarios.

22 0

Paper Graph

Unsupervised Continual Anomaly Detection with Contrastively-learned Prompt

Chengjie Wang, Kai Wu, Qiang Nie, Ying Chen, Bin-Bin Gao, Yong Liu, Jinbao Wang, Feng Zheng•Mon Jan 01 2024

A novel Unsupervised Continual Anomaly Detection framework called UCAD is introduced, which equips the UAD with continual learning capability through contrastively-learned prompts, and Structure-based Contrastive Learning (SCL) is designed with the Segment Anything Model to improve prompt learning and anomaly segmentation results.

44 0

Paper Graph

Adding a benchmark result helps the community track progress.