ncnn-yolov8-seg

This is a sample ncnn yolov8 object segment android project, it depends on ncnn library and opencv.

How to convert yolov8 model to ncnn model?

1 How to build and run

1.1 Configure ncnn

• Download ncnn-YYYYMMDD-android-vulkan.zip.
• Extract ncnn-YYYYMMDD-android-vulkan.zip into app/src/main/jni folder and change the ncnn_DIR path to yours in app/src/main/jni/CMakeLists.txt.

For example:ncnn-20221128-android-vulkan

1.2 Configure OpenCV

• Download opencv-mobile-XYZ-android.zip
• Extract opencv-mobile-XYZ-android.zip into app/src/main/jni and change the OpenCV_DIR path to yours in app/src/main/jni/CMakeLists.txt.

For example:opencv-mobile-4.6.0-android

1.3 Build and Install ncnn-yolov8-seg app

• Open this project with Android Studio, build it and enjoy!

2 Some notes

• Android ndk camera is used for best efficiency.
• Crash may happen on very old devices for lacking HAL3 camera interface.
• All models are manually modified to accept dynamic input shape.
• Most small models run slower on GPU than on CPU, this is common.
• FPS may be lower in dark environment because of longer camera exposure time.

3 Screenshot

fp16 cpu infer

fp32 cpu infer

4 Reference

• https://github.com/ultralytics/ultralytics
• https://github.com/Tencent/ncnn
• https://github.com/nihui/opencv-mobile
• https://github.com/nihui/ncnn-android-nanodet
• https://github.com/FeiGeChuanShu/ncnn-android-yolov8