# Mediapipe-android-pose-tracking

**Repository Path**: luo_zhi_cheng/Mediapipe-android-pose-tracking

## Basic Information

- **Project Name**: Mediapipe-android-pose-tracking
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 2
- **Forks**: 2
- **Created**: 2021-05-26
- **Last Updated**: 2023-01-10

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

This is an example of using MediaPipe AAR in Android Studio with Gradle.

The steps to build and use MediaPipe AAR is documented in MediaPipe's [android_archive_library.md](https://github.com/google/mediapipe/blob/master/docs/getting_started/android_archive_library.md). 

1. To control the output stream, you can modify https://github.com/google/mediapipe/blob/master/mediapipe/graphs/pose_tracking/pose_tracking_gpu.pbtxt to generate your own pose_tracking_gpu.binarypb, then overwrite it in the folder app/src/main/assets/

2. Google ML Kit Pose detection.
   > https://developers.google.com/ml-kit/vision/pose-detection

3. After my testing, the current Mediapipe pose detection model is more accurate than ML Kit even if Google said ML Kit also uses BlazePose.