search

ekosman / AnomalyDetectionCVPR2018-Pytorch

Pytorch version of - https://github.com/WaqasSultani/AnomalyDetectionCVPR2018
176 stars 55 forks source link

Doubt in Train_Annotation.txt #148

Closed prasaddev9 closed 1 year ago

prasaddev9 commented 1 year ago

Could you provide me the "pretrained_3d path" in "feature_extractor.py" of both c3d and i3d? Screenshot from 2022-12-30 17-26-47

ekosman commented 1 year ago

I3D is not supported yet. C3D weights are available in the README.

prasaddev9 commented 1 year ago

Thanks @ekosman.

I am trying to implement I3D network. After implementation, I'll share the code with you.

Btw I am having a doubt in "Train_Annotation.txt".

Screenshot from 2023-01-04 08-29-30

  1. What does it signifies in second column of each annotation file of Train_Annotation.txt [Screenshot attached] ?

  2. Also how can we create our own annotation ?

ekosman commented 1 year ago

They mean the length of the video. You can just append random numbers there. As I just took these annotations from the original repo, the code doesn't make use of it.

Thanks for asking. I will clarify this in the README.

Thanks @ekosman.

I am trying to implement I3D network. After implementation, I'll share the code with you.

Btw I am having a doubt in "Train_Annotation.txt".

Screenshot from 2023-01-04 08-29-30

1. What does it signifies in second column of each annotation file of Train_Annotation.txt [Screenshot attached] ?

2. Also how can we create our own annotation ?
ekosman commented 1 year ago

Unless there are further questions, I'll close this issue by 15/1