【ncnn android】算法移植（五）——DBFace移植

1. DBface簡介

• 爲什麼用DBface
  dbface，centerface原理都很簡單直接，而且可拓展性好

• 爲什麼用DBface
  最新的版本進行了模型backbone重設計，模型大小隻有1.4M，適合嵌入式設備

2. pth2onnx

pytorch 用1.3.0

git clone https://github.com/dlunion/DBFace
cd <dbface path>/train/small

# 修改trial_name
trial_name = "small-H-dense-wide64-UCBA-keep12-noext-ignoresmall2"

3. onnx2ncnn

參考上一篇博文編譯ncnn之後，在tools的文件夾下有onnx2ncnn的可執行文件。copy到onnx模型所在的文件夾。就可以轉換了。

./onnx2ncnn ./model.onnx dbface.param dbface.bin

報錯

Sigmoid not supported yet!
Sigmoid not supported yet!
Sigmoid not supported yet!
Sigmoid not supported yet!
Sigmoid not supported yet!
Sigmoid not supported yet!
Plugin not supported yet!
  # info={"scale_factor": [2, 2]}
  # name=Bilinear
Plugin not supported yet!
  # info={"scale_factor": [2, 2]}
  # name=Bilinear
Plugin not supported yet!
  # info={"scale_factor": [2, 2]}
  # name=Bilinear
Sigmoid not supported yet!
Exp not supported yet!

TO DO

• 拓展ncnn的layer，支持sigmoid，bilinear layer