Hey Guys i hope there are some people out here who worked with the image/video training for the mediapipe framework.
I am trying to feed the mediapipe framework with as much pictures as i have but i find it very exhausting to make pictures of people from every angle to get data from people.
I did not try the video training for the mediapipe yet, because i hate it to cut videos.
As a example i would like to train the exercise squad, so i need three states to have a squad
- standing
- squadDown
- squadUp
For every of those states i need a lot of training data to feed the AI otherwise it will not count my movement.
So my question to you guys is the following.
What is the best way to get as much pictures from movements without much time waste.
Making manuell pictures but i realisized how time comsuming this would be
Roni is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.