Self-trained Deep Ordinal Regression for End-to-End Video Anomaly Detection

[llien30]

INFO

Author

Guansong Pang, Cheng Yan, Chunhua Shen, Anton van den Hengel, Xiao Bai

Affiliation

Conference or Year

CVPR2020

Link

• arXiv
• [official GitHub]()

Abstract

• Adapt self-training ordinal regression to video anomaly detection
• Present novel end-to-end NNs based anomaly detection method -> Generates optimal anomaly scores -> Enables effective human-in-the-loop anomaly detection -> Easy and accurate localization of the identifies anomalies within the corresponding images

Proposed Method

Evaluation

AUC performance

Contribution

Discussion, Future Work

Comment

self-training の異常検出は初めて見た気がする

Date

2020/06/21