hxhh / ncnn_example

ncnn example: mask detection: anticonv face detection: retinaface&&mtcnn&&centerface, track: iou tracking, landmark: zqcnn, recognize: mobilefacenet classifier: mobilenet object detecter: mobilenetssd

ncnn_example:

I promise i'll update this project forever!

2020.03.10

1.add mask detect code

2.optimize the name of retinaface

3.update ncnn lib version to newest: 20200226

2020.02.23:

1. add vulkan option

2. fix the network name of mobilenetssd

2020.02.22:

1. split the model files from the project

2. add classifier && add object detecter

How to use?

1. download the models from baiduyun: baidu(code: w48b) and google

2. put models to directory: ncnn_example/data/models

3. compile the project:

>> cd ncnn_example && mkdir build && cd build && make -j3 

4. run the project:

>> cd src && ./face && ./object && ./classifier

5. result

face result:

mask result:

object result:

classifier result:

6. TODO:

• optimize the speed by using block strategy
• add yolo
• add pose
• refactor the project

7. references

https://github.com/Tencent/ncnn

https://github.com/zuoqing1988/ZQCNN

https://github.com/Charrin/RetinaFace-Cpp

https://github.com/deepinsight/insightface

https://github.com/Star-Clouds/centerface

https://github.com/seetafaceengine/SeetaFace2