/GCN-Anomaly-Detection

Source codes of our paper in CVPR 2019: Graph Convolutional Label Noise Cleaner: Train a Plug-and-play Action Classifier for Anomaly Detection

Primary LanguagePython

GCN-Anomaly-Detection

LICENSE

Our source codes for the following paper at CVPR 2019:

Graph Convolutional Label Noise Cleaner: Train a Plug-and-play Action Classifier for Anomaly Detection

@inproceedings{adgcn_cvpr19,
  author    = {Jia-Xing Zhong and
	       Nannan Li and
               Weijie Kong and
               Shan Liu and
               Thomas H. Li and
               Ge Li},
  title     = {Graph Convolutional Label Noise Cleaner: Train a Plug-and-play Action Classifier for Anomaly Detection},
  booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
  year      = {2019}
}

Our codes for feature extraction have been released! The training codes are also available as in the folder pygcn.