This repo contains the Pytorch implementation of our paper:
Weakly-supervised Video Anomaly Detection with Robust Temporal Feature Magnitude Learning
Please download the extracted I3d features for ShanghaiTech dataset from these two links:
[ShanghaiTech feature on Google dirve](https://drive.google.com/drive/folders/1L71Qa0gao6aLVhSjL0H-u2khmTRKcmQs?usp=sharing, https://drive.google.com/drive/folders/1z-CQPpVtTyfZyPKZdv2hZ-h2oMF6s8ep?usp=sharing)
Extracted I3d features for Ucf-Crime:
The following files need to be adapted in order to run the code on your own machine:
- Change the file paths to the download datasets above in
list/shanghai-i3d-test-10crop.list
andlist/shanghai-i3d-train-10crop.list
for ShanghaiTech, orlist/ucf-i3d.list
andlist/ucf-i3d-test.list
for Ucf-Crime. - Feel free to change the hyperparameters in
option.py
After the setup, simply run the following command:
python main.py