jimchenhub/new-caffe-segnet

Apply segnet in new version of caffe

New Caffe SegNet

To use some new features, I implement the SegNet in current version of Caffe.

Because Caffe has already have Batch Normalization Layer, I just implement Upsample Layer and Dense Image Data Layer.

Caffe

Caffe is a deep learning framework made with expression, speed, and modularity in mind. It is developed by the Berkeley Vision and Learning Center (BVLC) and community contributors.

Check out the project site for all the details like

• DIY Deep Learning for Vision with Caffe
• Tutorial Documentation
• BVLC reference models and the community model zoo
• Installation instructions

and step-by-step examples.

Please join the caffe-users group or gitter chat to ask questions and talk about methods and models. Framework development discussions and thorough bug reports are collected on Issues.

Happy brewing!

License and Citation

Caffe is released under the BSD 2-Clause license. The BVLC reference models are released for unrestricted use.

Please cite Caffe in your publications if it helps your research:

@article{jia2014caffe,
  Author = {Jia, Yangqing and Shelhamer, Evan and Donahue, Jeff and Karayev, Sergey and Long, Jonathan and Girshick, Ross and Guadarrama, Sergio and Darrell, Trevor},
  Journal = {arXiv preprint arXiv:1408.5093},
  Title = {Caffe: Convolutional Architecture for Fast Feature Embedding},
  Year = {2014}
}

new-caffe-segnet's Issues

bn_param layer appears to be missing

[libprotobuf ERROR google/protobuf/text_format.cc:274] Error parsing text-format caffe.NetParameter: 51:12: Message type "caffe.LayerParameter" has no field named "bn_param".

It looks like bn_layer.cpp and *.cu are missing from your src/layers folder

Do I need to change the bn layer to batch_normalization or whatever the newer version of caffe uses in the prototxt?

Missing .Doxyfile and error in Cuda.cmake

Missing .Doxyfile:

I had to manually add the .Doxyfile for it to compile correctly.

Error in Cuda.cmake:

I have cudnn version 5.1.3 installed, but kept getting the error that it has to be >3, so I had to edit the detect_cuDNN function to add

set(CUDNN_VERSION_MAJOR "5")

after the

if(NOT CUDNN_VERSION_MAJOR)
    set(CUDNN_VERSION "???")
else()
    set(CUDNN_VERSION "${CUDNN_VERSION_MAJOR}.${CUDNN_VERSION_MINOR}.${CUDNN_VERSION_PATCH}")
endif()

since I think it doesn't correctly get the version number from cudnn.h at the top of the function