https://github.com/glcas/facedemo
A demo of real-time face recognition on Android.
https://github.com/glcas/facedemo
camerax java mobilefacenet mtcnn ncnn
Last synced: 3 months ago
JSON representation
A demo of real-time face recognition on Android.
- Host: GitHub
- URL: https://github.com/glcas/facedemo
- Owner: glcas
- Created: 2020-08-10T10:21:44.000Z (about 5 years ago)
- Default Branch: master
- Last Pushed: 2022-03-05T15:37:31.000Z (over 3 years ago)
- Last Synced: 2025-04-08T19:34:57.229Z (6 months ago)
- Topics: camerax, java, mobilefacenet, mtcnn, ncnn
- Language: C
- Homepage:
- Size: 20 MB
- Stars: 15
- Watchers: 4
- Forks: 8
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# faceDemo
A demo of real-time face recognition on Android.
## Features
- Used [CameraX](https://developer.android.com/training/camerax) to implement the preview and get images for analysis.
- Used ncnn2mem tool to convert the model files to header files, making the models hardcoded. Thus the demo can work once installed.
- Support simultaneous multi-face recognition.
## References
[ncnn](https://github.com/Tencent/ncnn)
[mtcnn_ncnn](https://github.com/moli232777144/mtcnn_ncnn)
[mobilefacenet-V2](https://github.com/qidiso/mobilefacenet-V2)
[mobilefacenet-ncnn](https://github.com/liguiyuan/mobilefacenet-ncnn)
The CameraX provides image analysis use case, which produces images in YUV_420_888 format. The code [here](https://stackoverflow.com/questions/36212904/yuv-420-888-interpretation-on-samsung-galaxy-s7-camera2) can convert the images to Bitmap format.
## Screenshots
