README.md

Margin Learning Embedded Prediction for Video Anomaly Detection with A Few Anomalies, IJCAI 2019

Wen Liu*, Weixin Luo*, Zhengxin Li, Peilin Zhao, Shenghua Gao.

1. Installation (Anaconda with python3.6 installation is recommended)

pip install -r requirements.txt

2. Download datasets

Please manually download all datasets from avenue.tar.gz and shanghaitech.tar.gz and tar each tar.gz file, and move them in to data folder.

You can also download data from BaiduYun(https://pan.baidu.com/s/1j0TEt-2Dw3kcfdX-LCF0YQ) i9b3

3. Inference the pretrain model

Download the pre-trained models firstly, pretrains folder

• from OneDrive;
• from BaiduPan, ioov

and then, move the pretrains folder into data,mv pretrains data.

3.1 Inference with Only-Normal-Data Pretrained model

python inference.py  --dataset  avenue    \
          --prednet  cyclegan_convlstm    \
          --num_his  4                     \
          --label_level  normal            \
          --gpu      0                     \
          --interpolation  --snapshot_dir  ./data/pretrains/avenue/normal/checkpoints/model.ckpt-74000

3.2 Inference with Video-Annotated Pretrained model

python inference.py  --dataset  avenue    \
          --prednet  cyclegan_convlstm    \
          --num_his  4                     \
          --label_level  tune_video            \
          --gpu      0                     \
          --interpolation  --snapshot_dir  ./data/pretrains/avenue/tune_video/prednet_cyclegan_convlstm_folds_10_kth_1_/MARGIN_1.0_LAMBDA_1.0/model.ckpt-76000

3.3 Inference with Temporal-Annotated Pretrained model

python inference.py  --dataset  avenue    \
          --prednet  cyclegan_convlstm    \
          --num_his  4                     \
          --label_level  normal            \
          --gpu      0                     \
          --interpolation  --snapshot_dir  ./data/pretrains/avenue/temporal/prednet_cyclegan_convlstm_folds_10_kth_1_/MARGIN_1.0_LAMBDA_1.0/model.ckpt-77000

4. Training model with different settings from scratch

See more details in

4.1 only_normal_data;

4.2 video_annotation;

4.3 temporal_annotation.

Citation

@inproceedings{melp_2019,
  author    = {Wen Liu and
               Weixin Luo and
               Zhengxin Li and
               Peilin Zhao and
               Shenghua Gao},
  title     = {Margin Learning Embedded Prediction for Video Anomaly Detection with
               {A} Few Anomalies},
  booktitle = {Proceedings of the Twenty-Eighth International Joint Conference on
               Artificial Intelligence, {IJCAI} 2019, Macao, China, August 10-16,
               2019},
  pages     = {3023--3030},
  publisher = {ijcai.org},
  year      = {2019}
}