Face Detection with the Faster R-CNN

main site 

https://github.com/playerkk/face-py-faster-rcnn

download `faster-rcnn-models.tgz` at 

https://dl.dropboxusercontent.com/s/o6ii098bu51d139/faster_rcnn_models.tgz ref

All pre-trained models cannot be downloaded. All links to data are broken.

It seems like not working. caffe which is included to source git is not compiled. Looks like cuder version related problem but not sure.

cudnn_sigmoid_layer.cu(13): error: argument of type "cudnnActivationMode_t" is incompatible with parameter of type "cudnnActivationDescriptor_t"
         detected during instantiation of "void caffe::CuDNNSigmoidLayer<Dtype>::Forward_gpu(const std::vector<caffe::Blob<Dtype> *, std::allocator<caffe::Blob<Dtype> *>> &, const std::vector<caffe::Blob<Dtype> *, std::allocator<caffe::Blob<Dtype> *>> &) [with Dtype=float]”

Even after independent caffe installation, it does not work. Maybe it's of caffe version problem.

WARNING: Logging before InitGoogleLogging() is written to STDERR
W0510 14:35:05.531553 18577 _caffe.cpp:139] DEPRECATION WARNING - deprecated use of Python interface
W0510 14:35:05.531584 18577 _caffe.cpp:140] Use this instead (with the named "weights" parameter):
W0510 14:35:05.531587 18577 _caffe.cpp:142] Net(‘/face/face-py-faster-rcnn/models/pascal_voc/VGG16/faster_rcnn_alt_opt/faster_rcnn_test.pt', 1, weights='/face/face-py-faster-rcnn/data/faster_rcnn_models/VGG16_faster_rcnn_final.caffemodel')
[libprotobuf ERROR google/protobuf/text_format.cc:274] Error parsing text-format caffe.NetParameter: 350:21: Message type "caffe.LayerParameter" has no field named "roi_pooling_param".
F0510 14:35:05.532542 18577 upgrade_proto.cpp:88] Check failed: ReadProtoFromTextFile(param_file, param) Failed to parse NetParameter file: /face/face-py-faster-rcnn/models/pascal_voc/VGG16/faster_rcnn_alt_opt/faster_rcnn_test.pt
*** Check failure stack trace: ***
Aborted (core dumped)

https://arxiv.org/pdf/1701.08289.pdf 이 논문 구현체인가.. 아닌것 같은데.. 확인필요.

blog comments powered by Disqus