Realtime Face Detection with Firebase ML Kit
summary
Shows a working realtime Snapchat-style filter built with Firebase ML Kit on-device face detection, Fotoapparat camera view and RxPermissions, with full open-source code and a codelab. The detector returns bounding box, landmarks, Euler angles, tracking IDs and optional contours expensive output; enabling contours downloads a TensorFlow model and slows processing. Uses detector options (choose minFaceSize 10% to ignore small faces), processes camera preview bitmaps, avoids double-rendering, drops frames and reduces preview resolution for smooth realtime performance — demo runs well when faces are close.
AI-written summary · about
chapters (9)
- 0:00Introduction and background
- 2:08Project roadmap
- 5:17Firebase ML theory
- 11:05Understanding face data
- 14:35Setting up the camera
- 19:26Implementing face detection
- 24:28Building custom graphics
- 29:00Integration and examples
- 35:42Real-time performance tips
related talks
30mMachine Learning in Android Apps with ML Kit
34mIt’s an ML full world! MLKit for Android Devs
21mLive Coding: Adding Machine Learning to Android Things in 10 Mins
32mSnapchat Filters with Flutter's CustomPainter and MLKit
29mPushing Machine Learning into your app When TensorFlow and MLKit rule the world
39mNow smile (also in 3D)! Exploring AR, ML and Camera related APIs on Android
30mMobile applications that can see
21mLeveraging Python for Real-Time Image processing