ncnn_106landmarks:

I promise i'll update this project forever!

1. add landmark

(1). the original model comes from repositories:

https://github.com/zuoqing1988/ZQCNN by zuoqing

(2). the process of converting mxnet model to ncnn formate is:

zuoqing1988/ZQCNN#50

(3). new landmark model: landmark_big.param && landmark_big.bin has also uploaded, input size: 48x48

you can change the model name in the test code to test the model:

change "landmark.param", "landmark.bin" to "landmark_big.param" && "landmark_big.bin"

(4). add input size: 96x96 model:

you need to change the input size 48 x 48 to 96 x 96:

ncnn::Mat ncnn_in = ncnn::Mat::from_pixels_resize(img_src.data,

ncnn::Mat::PIXEL_BGR, img_src.cols, img_src.rows, 48, 48);

To:

ncnn::Mat ncnn_in = ncnn::Mat::from_pixels_resize(img_src.data,

ncnn::Mat::PIXEL_BGR, img_src.cols, img_src.rows, 96, 96);

(5).add input size: 112x112 model:

the most effect model.

2. add retinaface detection

the code refer to the repositories:

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

model comes from:

https://github.com/Charrin/RetinaFace-Cpp/tree/master/Demo/ncnn/models

result:

3.add jetson nano project based on vulkan

(1). build vulkan ncnn:

https://github.com/Tencent/ncnn/wiki/FAQ-ncnn-vulkan

(2). build the project:

>mkdir build && cd build && cmake .. && make -j3

>./main

4.add mobilefacenet

5.use openmp to optimize for loops

test result:

do not use vulkan:

' start face detect. 4 faces detected. start keypoints extract. keypoints extract end. start keypoints extract. keypoints extract end. start keypoints extract. keypoints extract end. start keypoints extract. keypoints extract end. time cost: 137.495ms '

use vulkan:

' [0 NVIDIA Tegra X1 (nvgpu)] queueC=0[16] queueT=0[16] memU=2 memDL=2 memHV=2 [0 NVIDIA Tegra X1 (nvgpu)] fp16p=1 fp16s=1 fp16a=0 int8s=1 int8a=0 start face detect. 4 faces detected. start keypoints extract. keypoints extract end. start keypoints extract. keypoints extract end. start keypoints extract. keypoints extract end. start keypoints extract. keypoints extract end. time cost: 553.328ms '

why so strange?

6.add face alignment interface

the interface refer to the project insightface:

https://github.com/deepinsight/insightface/tree/master/cpp-align

and the issue:

deepinsight/insightface#333

7.refactor the project

reduce the coupling of modules

8.add mtcnn

You only need to change line 14 of face_engine.cpp to change face detector.

For Example:

detector_(new Mtcnn()) To detector_(new RetinaFace())

9.optimize the structure of the project

TODO:

-[x]. add face recognize database

-[x]. add face track