R-FCN + ResNet + Python + 训练自己的模型

安装MS Caffe

假设你已经装好了CUDA，cudnn，caffe等等。 这一步感觉和faster rcnn很像，都需要独特版本的caffe。这里的版本就是Microsoft的caffe。

pip install cython pip install easydict apt-get install
python-opencv
git clone https://github.com/Orpine/py-R-FCN.git
cd py-R-FCN

进入lib文件夹，然后：

make

接着进入caffe目录：（32核用起来果然很爽）

make -j32 && make pycaffe

这里的Makefile.config需要注意一下，每个人的都有一点点不一样，比如我就没有加入opencv。这里没写好的话就会有不少错误。

## Refer to http://caffe.berkeleyvision.org/installation.html# Contributions simplifying and improving our build system are welcome!# cuDNN acceleration switch (uncomment to build with cuDNN).USE_CUDNN := 1# CPU-only switch (uncomment to build without GPU support).# CPU_ONLY := 1# uncomment to disable IO dependencies and corresponding data layers# USE_OPENCV := 0# USE_LEVELDB := 0# USE_LMDB := 0# uncomment to allow MDB_NOLOCK when reading LMDB files (only if necessary)#       You should not set this flag if you will be reading LMDBs with any#       possibility of simultaneous read and write# ALLOW_LMDB_NOLOCK := 1# Uncomment if you're using OpenCV 3# OPENCV_VERSION := 3# To customize your choice of compiler, uncomment and set the following.# CUDA directory contains bin/ and lib/ directories that we need.CUDA_DIR := /usr/local/cuda# On Ubuntu 14.04, if cuda tools are installed via# "sudo apt-get install nvidia-cuda-toolkit" then use this instead:# CUDA_DIR := /usr# CUDA architecture setting: going with all of them.# For CUDA < 6.0, comment the *_50 lines for compatibility.CUDA_ARCH := -gencode arch=compute_20,code=sm_20 \                -gencode arch=compute_20,code=sm_21 \                -gencode arch=compute_30,code=sm_30 \                -gencode arch=compute_35,code=sm_35 \                -gencode arch=compute_50,code=sm_50 \                -gencode arch=compute_50,code=compute_50# BLAS choice:# atlas for ATLAS (default)# mkl for MKL# open for OpenBlasBLAS := atlas# Custom (MKL/ATLAS/OpenBLAS) include and lib directories.# Leave commented to accept the defaults for your choice of BLAS# open for OpenBlasBLAS := atlas# Custom (MKL/ATLAS/OpenBLAS) include and lib directories.# Leave commented to accept the defaults for your choice of BLAS# (which should work)!# BLAS_INCLUDE := /path/to/your/blas# BLAS_LIB := /path/to/your/blas# Homebrew puts openblas in a directory that is not on the standard search path# BLAS_INCLUDE := $(shell brew --prefix openblas)/include# BLAS_LIB := $(shell brew --prefix openblas)/lib# This is required only if you will compile the matlab interface.# MATLAB directory should contain the mex binary in /bin.# MATLAB_DIR := /usr/local# MATLAB_DIR := /Applications/MATLAB_R2012b.app# NOTE: this is required only if you will compile the python interface.# We need to be able to find Python.h and numpy/arrayobject.h.PYTHON_INCLUDE := /usr/include/python2.7 \                /usr/lib/python2.7/dist-packages/numpy/core/include# Anaconda Python distribution is quite popular. Include path:# Uncomment to support layers written in Python (will link against Python libs)WITH_PYTHON_LAYER := 1# Whatever else you find you need goes here.INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include /usr/lib/x86_64-linux-gnu/hdf5/serial/includeLIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib /usr/lib/x86_64-linux-gnu/hdf5/serial# If Homebrew is installed at a non standard location (for example your home directory) and you use it for general dependencies# INCLUDE_DIRS += $(shell brew --prefix)/include# LIBRARY_DIRS += $(shell brew --prefix)/lib# Uncomment to use `pkg-config` to specify OpenCV library paths.# (Usually not necessary -- OpenCV libraries are normally installed in one of the above $LIBRARY_DIRS.)# USE_PKG_CONFIG := 1# N.B. both build and distribute dirs are cleared on `make clean`BUILD_DIR := buildDISTRIBUTE_DIR := distribute# Uncomment for debugging. Does not work on OSX due to https://github.com/BVLC/caffe/issues/171# DEBUG := 1# The ID of the GPU that 'make runtest' will use to run unit tests.TEST_GPUID := 0# enable pretty build (comment to see full commands)Q ?= @       

测试Demo

下载已经训练好的模型。
点击这里： https://1drv.ms/u/s!AoN7vygOjLIQqUWHpY67oaC7mopf

然后下载之后放进去：

$py-R-FCN/data/rfcn_models/resnet50_rfcn_final.caffemodel

$py-R-FCN/data/rfcn_models/resnet101_rfcn_final.caffemodel

运行：

./tools/demo_rfcn.py –net ResNet-50

效果就是：

训练自己的模型的准备工作

在data目录下，下载VOC的数据集。
1、下载

wget http://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCtrainval_06-Nov-2007.tarwget http://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCtest_06-Nov-2007.tarwget http://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCdevkit_08-Jun-2007.tarwget http://host.robots.ox.ac.uk/pascal/VOC/voc2012/VOCtrainval_11-May-2012.tar

2、解压：

tar xvf VOCtrainval_06-Nov-2007.tartar xvf VOCtest_06-Nov-2007.tartar xvf VOCdevkit_08-Jun-2007.tartar xvf VOCtrainval_11-May-2012.tar

3、整理：
把下面的情况：

$VOCdevkit/                           # development kit$VOCdevkit/VOCcode/                   # VOC utility code$VOCdevkit/VOC2007                    # image sets, annotations, etc.$VOCdevkit/VOC2012                    # image sets, annotations, etc.

弄成： 多一个0712

$VOCdevkit/                           # development kit  $VOCdevkit/VOCcode/                   # VOC utility code  $VOCdevkit/VOC2007                    # image sets, annotations, etc.  $VOCdevkit/VOC2012                    # image sets, annotations, etc.  $VOCdevkit/VOC0712                    # you just created this folder

PS：在这里需要注意把训练的train.sh里面的TRAIN_IMDB改成如下形式。

case $DATASET in    pascal_voc)      TRAIN_IMDB="voc_0712_trainval"      TEST_IMDB="voc_0712_test"      PT_DIR="pascal_voc"      ITERS=110000  

4、链接：

cd $RFCN_ROOT/dataln -s $VOCdevkit VOCdevkit0712

5、下载事先训练好的imageNet的rest50，100的模型。

ImageNet-pre-trained ResNet-50 and ResNet-100 model 

下载地址：链接：http://pan.baidu.com/s/1slRHD0L 密码：r3ki

下载之后放在：/data/imagenet_models目录下。

训练

需要修改网络接口，我的输出是4类，那么在需要改不少东西。具体修改参考： http://blog.csdn.net/sinat_30071459/article/details/53202977
http://blog.csdn.net/xzzppp/article/details/52036794
这篇博客已经写得非常详细。

接着就使用 ./experiments/scripts/rfcn_end2end_ohem.sh 0 ResNet-50 pascal_voc 训练。

$RFCN_ROOT/experiments/scripts里还有一些其他的训练方法，也可以测试一下（经过上面的修改，无ohem的end2end训练也改好了，其他训练方法修改的过程差不多）。

将训练得到的模型

($RFCN_ROOT/output/rfcn_end2end_ohem/voc_0712_trainval

里最后的caffemodel)拷贝到

$RFCN_ROOT/data/rfcn_models

下，然后打开

$RFCN_ROOT/tools/demo_rfcn.py

，将CLASSES修改成你的标签，NETS修改成你的model，im_names修改成你的测试图片(放在data/demo下),最后：

./tools/demo_rfcn.py –net ResNet-50

看一下最后的结果：

参考链接

1、http://blog.csdn.net/sinat_30071459/article/details/53202977
2、https://github.com/Orpine/py-R-FCN