PyAnomaly is the open-source tool for anomaly detection, which provides a tool for researchers and engineers to accelerate their study and development.

More introduction of the toolbox is in INTRO

!!!!Attention: Please use the stable branch, I'm still working the master branch to build the PyAnomlay 2.0. Thanks for your attention.

• [Dec 2020] Add the API document, please refer to API for more information
• [Nov 2020] Change the methods to get model, make it much easier to use
• [Jun 2020] The repo is open!!

Note: For the detailed changing logs, please refer to the CHANGE

Please refer to the INSTALL

Specifically usage please refer to USAGE

The introduction of some useful tools is in TOOLS

The API document is in API

This project is released under the Apache 2.0 license.

@inproceedings{pyanomaly,
author = {Cheng, Yuhao and Liu, Wu and Duan, Pengrui and Liu, Jingen and Mei, Tao},
title = {PyAnomaly: A Pytorch-Based Toolkit for Video Anomaly Detection},
year = {2020},
publisher = {Association for Computing Machinery},
booktitle = {Proceedings of the 28th ACM International Conference on Multimedia},
pages = {4473–4476},
keywords = {open-source, toolkit, video anomaly detection},
location = {Seattle, WA, USA},
}