I would like to thank you for your interesting work. How to identify the causes of violence？ Thanks again for your interesting work.

[JimLee1996]

please see https://github.com/JimLee1996/AVSS2019/tree/master/src/Demo

[lawoyiba]

Is it through this line of code? def predict(self, x): x = x.to(self.device)

need N x C x T x H x W

    if len(x.shape) < 5:
        x = x.unsqueeze(0)
    y = self.net(x)

    _, results = y.topk(1, 1, True)

    labels = ['normal' if x else 'violent' for x in results]

    return labels

[JimLee1996]

Yep

[lawoyiba]

Thank you for your reply. I read your paper. What are the methods used to distinguish violent behavior from normal behavior? Thank you again for your work

[JimLee1996]

We use deep learning methods to do this job. 3DCNNs are proved to be effective on learning spatio-temporal features. Here we introduce DenseNet architecture to better learn the motion pattern of violent behavior. The outputs of model are logits of classes (in this task, violence and non-violence)

[lawoyiba]

Thank you for your reply, After the video passes the DenseNet model, it will identify whether there is violence in the video.

[JimLee1996]

Use softmax. Please refer to https://github.com/JimLee1996/AVSS2019/issues/2.

[lawoyiba]

Thank you for your reply