Beluga Journey
Reproduce Akanazawa Motion Reconstruction: Day 2 ish

Reproduce Akanazawa Motion Reconstruction: Day 2 ish

Claire Chen's photo
Claire Chen
·

3 min read

OpenPose

Motion Reconstruction

Progress:

  • git clone protobuf build and install
sudo apt-get install autoconf automake libtool curl make g++ unzip -y
git clone https://github.com/google/protobuf.git
cd protobuf
git submodule update --init --recursive
./autogen.sh
./configure
make
make check
sudo make install
sudo ldconfig
  • install cmake-gui, boost, atlas, + other dependencies
sudo apt-get install cmake-qt-gui
sudo apt-get install libboost-all-dev
sudo apt-get install libatlas-base-dev
sudo apt-get install libgoogle-glog-dev
sudo apt-get install libopencv-dev
  • build openpose 
make -j`nproc`
  • demo vids (imagine this picture is a video xd)
./build/examples/openpose/openpose.bin --video examples/media/video.avi

openposedemo.png

Fixed Issues:

  • Fixed openpose directory to my local openpose repo from
-kOpenPose = '/scratch1/storage/git_repos/openpose
  • Deprecated command line
ERROR: unknown command line flag 'write_keypoint_json'

-write_json (Directory to write OpenPose output in JSON format. It includes
      body, hand, and face pose keypoints (2-D and 3-D), as well as pose
      candidates (if `--part_candidates` enabled).) type: string default: ""
-write_keypoint ((Deprecated, use `write_json`) Directory to write the
      people pose keypoint data. Set format with `write_keypoint_format`.)
      type: string default: ""

-    cmd_base = '%s/build/examples/openpose/openpose.bin --video %%s --write_keypoint_json %%s --net_resolution "1312x736" --scale_number 4 --scale_gap 0.25 --write_images %%s --write_images_format jpg' % (
+    cmd_base = '%s/build/examples/openpose/openpose.bin --video %%s --write_json %%s --net_resolution "1312x736" --scale_number 4 --scale_gap 0.25 --write_images %%s --write_images_format jpg' % (

Outstanding Issues:

kOpenPoseModel = '/scratch1/storage/git_repos/Realtime_Multi-Person_Pose_Estimation/aj_finetuned_models_170k/'
  • Openpose crash when detecting face and hand
./build/examples/openpose/openpose.bin --video examples/media/video.avi --face --hand
Starting OpenPose demo...
Configuring OpenPose...
Starting thread(s)...
Auto-detecting all available GPUs... Detected 1 GPU(s), using 1 of them starting at GPU 0.
F0306 23:24:35.079802 44493 cudnn_conv_layer.cpp:53] Check failed: status == CUDNN_STATUS_SUCCESS (1 vs. 0)  CUDNN_STATUS_NOT_INITIALIZED
*** Check failure stack trace: ***
    @     0x7f0589c721c3  google::LogMessage::Fail()
    @     0x7f0589c7725b  google::LogMessage::SendToLog()
    @     0x7f0589c71ebf  google::LogMessage::Flush()
    @     0x7f0589c726ef  google::LogMessageFatal::~LogMessageFatal()
    @     0x7f05898d3d55  caffe::CuDNNConvolutionLayer<>::LayerSetUp()
    @     0x7f05899b61ad  caffe::Net<>::Init()
    @     0x7f05899b85b0  caffe::Net<>::Net()
    @     0x7f058a32b1ec  op::NetCaffe::initializationOnThread()
    @     0x7f058a2d65a8  op::FaceExtractorCaffe::netInitializationOnThread()
    @     0x7f058a2d7ef5  op::FaceExtractorNet::initializationOnThread()
    @     0x7f058a3710e7  op::Worker<>::initializationOnThreadNoException()
    @     0x7f058a371238  op::SubThread<>::initializationOnThread()
    @     0x7f058a372a48  op::Thread<>::initializationOnThread()
    @     0x7f058a37601c  op::Thread<>::threadFunction()
    @     0x7f0589fb1d84  (unknown)
    @     0x7f0589cb4609  start_thread
    @     0x7f0589df0293  clone
Aborted (core dumped)
TensorFlowMachine Learning