zheqiushui / faster-retinaface-cpp-mxnet

Faster-RetinaFace-Cpp-mxnet

this repositories retinaface is faster than my the other repositories,and faster than Official demo!!

Special mention!!!

RetinaFace-Cpp-caffe-version:this programing is the best with retinaface code by cpp

dwjlw1314:the main code have change to mxnet was provide by this brother.thank!

Model from deepinsight/insightface/RetinaFace

deepinsight:

Pretrained Model: RetinaFace-R50 (baidu cloud or dropbox) is a medium size model with ResNet50 backbone. It can output face bounding boxes and five facial landmarks in a single forward pass.

WiderFace validation mAP: Easy 96.5, Medium 95.6, Hard 90.4.

To avoid the confliction with the WiderFace Challenge (ICCV 2019), we postpone the release time of our best model.

Third-party Models

yangfly:

RetinaFace-MobileNet0.25 (baidu cloud). WiderFace validation mAP: Hard 82.5. (model size: 1.68Mb)

Depend

opencv cpp lib

mxnet cpp lib

cuda 10.0

cudnn 7.5

faster-retinaface-cpp-mxnet's Issues

Can you help for CMake or Make file

Can you help for compiling ? Cmake or Make file could be super

compiling issues

I have below errors:

../Mx-mtcnn/Faster_Retinaface.h:240:37: error: use of undeclared identifier 'anchor_num2'; did you mean 'anchor_num'? assert(cls.GetShape()[1] == anchor_num2); ^~~~~~~~~~~ anchor_num

../Mx-mtcnn/Faster_Retinaface.h:400:33: warning: comparison of integers of different signs: 'int' and 'size_type' (aka 'unsigned long') [-Wsign-compare] for (int _idx = 0; _idx < _feat_stride_fpn.size(); _idx++) ~~~~ ^ ~~~~~~~~~~~~~~~~~~~~~~~