Improving unsupervised anomaly localization by applying multi-scale memories to autoencoders

12/21/2020
by   Yifei Yang, et al.
8

Autoencoder and its variants have been widely applicated in anomaly detection.The previous work memory-augmented deep autoencoder proposed memorizing normality to detect anomaly, however it neglects the feature discrepancy between different resolution scales, therefore we introduce multi-scale memories to record scale-specific features and multi-scale attention fuser between the encoding and decoding module of the autoencoder for anomaly detection, namely MMAE.MMAE updates slots at corresponding resolution scale as prototype features during unsupervised learning. For anomaly detection, we accomplish anomaly removal by replacing the original encoded image features at each scale with most relevant prototype features,and fuse these features before feeding to the decoding module to reconstruct image. Experimental results on various datasets testify that our MMAE successfully removes anomalies at different scales and performs favorably on several datasets compared to similar reconstruction-based methods.

READ FULL TEXT

page 2

page 6

research
04/26/2021

ODDObjects: A Framework for Multiclass Unsupervised Anomaly Detection on Masked Objects

This paper presents a novel framework for unsupervised anomaly detection...
research
12/12/2022

Multi-scale Feature Imitation for Unsupervised Anomaly Localization

The unsupervised anomaly localization task faces the challenge of missin...
research
05/08/2022

Network Traffic Anomaly Detection Method Based on Multi scale Residual Feature

To address the problem that traditional network traffic anomaly detectio...
research
11/18/2022

Reference-Based Autoencoder for Surface Defect Detection

Due to the extreme imbalance in the number of normal data and abnormal d...
research
06/17/2023

Multi-scale Spatial-temporal Interaction Network for Video Anomaly Detection

Video anomaly detection (VAD) is an essential yet challenge task in sign...
research
02/24/2021

Automatic Feature Extraction for Heartbeat Anomaly Detection

We focus on automatic feature extraction for raw audio heartbeat sounds,...
research
08/08/2022

Clear Memory-Augmented Auto-Encoder for Surface Defect Detection

In surface defect detection, due to the extreme imbalance in the number ...

Please sign up or login with your details

Forgot password? Click here to reset