Applying computer vision tasks to Android apps - ML on Android with MediaPipe

Learn how to apply image classification, object detection, and hand gesture recognition into your Android app. Discover how to implement Machine Learning capabilities from MediaPipe for still images, videos, and a camera feed into your Android apps, which gives you the ability to add a lot of really cool and useful features to your apps without a ton of development overhead.

Resources:

#computervision #ML #Android

[Deeko]

The content is great, but the Google Advocate team really needs to provide some basic Android source code templates to get going so we can sample the features quickly.

[BaoQuocGuyen]

Good new skills for me! Thanks Google!

[banzai]

Fantastic Google Developers! It would be great to support Diffusers model (if possible)

[satheeshk]

I have fixed this alignment issue applying width and height different for for gallery and as well as for live feed, if i get a reply i will raise pr to mediapipe main branch

[pinocchio]

We need the source so that we can implement it well.

[megamind]

is there a way to use h5 models instead of task models?

[bindars]

is "gesture_recognizer.task" model included in the mediapipe itself?

[hinai]

When can there be an example of a pose classifier ？I need it.

[chin_]

Where can I find android code described in video?