nomorewzx/DORN_docker_setup

1. pull docker

   docker pull tensorflow/tensorflow:1.12.0-gpu

2. start docker

   docker run -it tensorflow/tensorflow:1.12.0-gpu /bin/bash

3. install libs on ubuntu

    apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libhdf5-serial-dev protobuf-compiler -y
    apt-get install --no-install-recommends libboost-all-dev -y
    apt-get install apt-get install cuda-cublas-dev-9-0 -y
    apt-get install libopenblas-dev -y
    apt-get install python-dev
    apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev -y
    apt-get install python-opencv -y
    pip install scikit-image
   

4. copy ./Makefile.config to {PATH_TO_DORN}/DORN/caffe/

   cp ./Makefile.config $PATH_TO_DORN/DORN/caffe/

5. make soft link of lib*.so

   ln -sv  /usr/lib/x86_64-linux-gnu/libhdf5_serial.so /usr/lib/x86_64-linux-gnu/libhdf5.so
   ln -sv  /usr/lib/x86_64-linux-gnu/libhdf5_serial_hl.so /usr/lib/x86_64-linux-gnu/libhdf5_hl.so
   

6. make caffe

    cd $PATH_TO_DORN/DORN/caffe/
    make all
    make test
    make runtest
   

7. export caffe/python to PYTHONPATH

    export PYTHONPATH = $PATH_TO_DORN/DORN/caffe/python
   

8. Download KITTI pretrained model from https://github.com/hufu6371/DORN, put model into $PATH_TO_DORN/DORN/models/

9. Run KITTI demo. Comment the GPU mode code. See ./demo_kitti.py in this repo, line 11 and line 12

    cd $PATH_TO_DORN/DORN/
    python demo_kitti.py --filename=./data/KITTI/demo_01.png --outputroot=./result/KITTI/
   

   check result in ./result/KITTI