Activity-Recognition
Activity-Recognition copied to clipboard
extract_frames.sh
when executing extract_frames.sh,did i need to create new folders in advance?
I've updated the repo file structure, you can now run it without having to make folders 1] ./extract_frames.sh 20 [for 20fps] 2] python3 Resize.py 3] python3 Optical_Flow_DualTVL1.py 4] python3 CNN_extract_Flow.py 5] python3 CNN_extract_RGB.py
After this comes the LSTM which you'll have to configure acc to your dataset.
PS: Thanks for pointing out the error, I forgot to update the repo
@ed-word thanks for your reply!by the way,i have a question that what does' Labels/X.csv' have?
It has the Y labels for the dataset. There are 2 columns one contains the path to the video, the second column has the Y label for that video
Have you tried on UCF-101 datasets?
I have to. Someday if I get time I will.