Home

Awesome

中文说明

:rocket: If it helps you, click a star! :star:

Ncnn deployment on mobile,support:YOLOv5s,YOLOv4-tiny,MobileNetV2-YOLOv3-nano,Simple-Pose,Yolact,ChineseOCR-lite,ENet,Landmark106,DBFace,MBNv2-FCN and MBNv3-Seg-small on camera.

iOS:

Android:

iOS

Select the model to be tested directly on the interface.

Android

Select the model to be tested directly on the interface.

Models

modelandroidiOSfromother
YOLOv5syesyesGithubTNN
YOLOv4-tinyyesyesGithub
YOLOv3-nanoyesyesGithub
YOLOv5s_custom_opyesyeszhihu
NanoDetyesyesGithubTNN MNN
YOLO-Fastest-xlyesyesGithub
Simple-PoseyesyesGithub
YolactyesyesGithub zhihu
ChineseOCR_liteyesyesGithub zhihu
ENetbugcancelGithub
Landmark106yesyesGithub
DBFaceyesyesGithub
MBNv2-FCNyesyesGithub
MBNv3-Seg-smallyesyesGithub
Light_OpenPoseyesyesGithub

iOS:

Android:

This project is more about practicing the use and deployment of various models, without too much processing in terms of speed. If you have requirements for speed, you can directly obtain data such as YUV for direct input or use methods such as texture and opengl to achieve data input, reducing intermediate data transmission and conversion.

Convert locally(Will not upload model): xxxx -> ncnn

Minimal OpenCV:opencv-mobile

:art: Screenshot<br/>

AndroidiOS
<img width="324" height="145" src="./Screenshots/Android_CPU_or_GPU.jpg"/><img width="320" height="166" src="./Screenshots/iOS_CPU_or_GPU.jpg"/>

Android

mbnv2-yolov3-nanoyolov4-tinyyolov5s
<img width="270" height="500" src="./Screenshots/Android_Meizu16x_mobilenetv2_yolov3_nano.jpg"/><img width="270" height="500" src="./Screenshots/Android_Meizu16x_yolov4_tiny.jpg"/><img width="270" height="500" src="./Screenshots/Android_Meizu16x_yolov5s.jpg"/>
simple_poseyolactchineseocr_lite_01
<img width="270" height="500" src="./Screenshots/Android_Meizu16x_simple_pose.jpg"/><img width="270" height="500" src="./Screenshots/Android_Meizu16x_yolact.jpg"/><img width="270" height="500" src="./Screenshots/Android_meizu16x_chineseocr_lite_01.jpg"/>
chineseocr_lite_02ENetyoloface500k-landmark106
<img width="270" height="500" src="./Screenshots/Android_meizu16x_chineseocr_lite_02.jpg"/><img width="270" height="500" src="./Screenshots/Android_Meizu16x_ENet.jpg"/><img width="270" height="500" src="./Screenshots/Android_Meizu16x_yoloface500k-landmark106.jpg"/>
dbfacembnv2_fcnmbnv3_seg_small
<img width="270" height="500" src="./Screenshots/Android_Meizu16x_dbface.jpg"/><img width="270" height="500" src="./Screenshots/Android_Meizu16x_MBNFCN.jpg"/><img width="270" height="500" src="./Screenshots/Android_Meizu16x_MBNV3_Seg_small.jpg"/>
yolov5s_custom_opnanodetyolo-fastest-xl
<img width="270" height="500" src="./Screenshots/Android_meizu16x_yolov5s_custom_layer.jpg"/><img width="270" height="500" src="./Screenshots/Android_meizu16x_nanodet.jpg"/><img width="270" height="500" src="./Screenshots/Android_meizu16x_yolo_fastest_xl.jpg"/>
light_openpose
<img width="270" height="500" src="./Screenshots/Android_Meizu16x_lightopenpose.jpg"/>

iOS

mbnv2-yolov3-nanoyolov4-tinyyolov5s
<img width="270" height="480" src="./Screenshots/iOS_iPhone6sp_mobilenetv2_yolov3_nano.jpg"/><img width="270" height="480" src="./Screenshots/iOS_iPhone6sp_yolov4_tiny.jpg"/><img width="270" height="480" src="./Screenshots/iOS_iPhone6sp_yolov5s.jpg"/>
yolov5s_custom_opnanodetyolo-fastest-xl
<img width="270" height="480" src="./Screenshots/iOS_iPhone6sp_yolov5s_custom_op.jpg"/><img width="270" height="480" src="./Screenshots/iOS_iPhone6sp_nanodet.jpg"/><img width="270" height="480" src="./Screenshots/iOS_iPhone6sp_yolo_fastest_xl.jpg"/>
mbnv2_fcnmbnv3_seg_smallsimple_pose
<img width="270" height="480" src="./Screenshots/iOS_iPhone6sp_mobilenetv2_fcn.jpg"/><img width="270" height="480" src="./Screenshots/iOS_iPhone6sp_mbnv3_segmentation_small.jpg"/><img width="270" height="480" src="./Screenshots/iOS_iPhone6sp_simple_pose.jpg"/>
chineseocr_lite_01chineseocr_lite_02light_openpose
<img width="270" height="480" src="./Screenshots/iOS_iPhone6sp_chineseocr_lite_01.jpg"/><img width="270" height="480" src="./Screenshots/iOS_iPhone6sp_chineseocr_lite_02.jpg"/><img width="270" height="480" src="./Screenshots/iOS_iPhone6sp_lightopenpose.jpg"/>
yolactyoloface500k-landmark106dbface
<img width="270" height="480" src="./Screenshots/iOS_iPhone6sp_yolact.jpg"/><img width="270" height="480" src="./Screenshots/iOS_iPhone6sp_yoloface500k_landmark106.jpg"/><img width="270" height="480" src="./Screenshots/iOS_iPhone6sp_dbface.jpg"/>

Thanks:<br/>