This application explores the functionality of some of Google’s Mediapipe Machine Learning solutions, viz:
- Hand Tracking
- Pose Estimation
- Face Detection
- Face Mesh
StreamLit is used to create the Web Graphical User Interface (GUI). This web app was inspired by the awesome YouTube tutorials created by:
Do check out their channels and websites for more informative and exciting Machine Learning & Computer Vision tutorials.
Demo
The app was been deployed on Streamlit. You can check it out here.
2 Likes
Thanks @Outsiders17711 for sharing your app with us! 
I wanted to check it out yet I currently have an error:
Are you aware of it?
Best,
Charly
Hello @Charly_Wargnier, thanks for reaching out.
I have noticed that if the app stays idle for a few hours, the page gives that error. And I don’t see anything related in the log when I check. Once I reboot the app, everything works perfect.
This is my first time deploying an app on Streamlit, so I don’t know if I did something wrong while deploying.
I have rebooted the app. You can check it out.
Thanks.
Fabulous, it looks great and I’ll give it a whirl! 
Best,
Charly
1 Like
i just open your app, and got this when i test mediapipe module
and when I changed to another page the same thing appeared
Hi,
It seems there is an issue with Streamlit that’s causing it to ignore the pinned version (0.86.0) I specified in my requirements.txt file. I have seen other threads with similar issues here and here.
Unfortunately, this is an old app and I haven’t updated the code in a while and don’t intend to. However, I’m currently trying out the solutions I found to see if any fixes this particular issue. I will keep you updated.
Cheers!
Hi @Dimas_Rizky_Wibowo,
Unfortunately, the issue is bigger than I thought. My Streamlit and Mediapipe versions are outdated, and updating their versions breaks the code in several places. I tried fixing the Streamlit issues, but when I found out that Mediapipe was also broken, I had to cut my losses.
As I said, this was an old project and I just don’t have the time to get it working right now. If you’re willing to do so, I created a branch of the repository to hold the updated Streamlit code.
Otherwise, I intend to leave the webapp as it is. Cheers!
I built an app which also uses Mediapipe for live (webcam) face mesh, pose detection and face detection:
1 Like
Hei @daltunay,
This looks awesome. The Mediapipe modules are better integrated than mine. The UI looks cleaner too! You can bet I will steal some of your code to update mine.
Thanks for sharing! Cheers!
2 Likes
Thank you @Outsiders17711! feel free to use my code, for sure 
1 Like
interesting, congratulations
1 Like
