Real-time Directional Segmentation from Kinect RGB-D Stream

This package implements real-time temporally consistent directional segmentation from Kinect RGB-D streams. It relies on the dpMMlowVar library for the actual implementation of the DDP-vMF-means algorithm. The approach and the algorithm is described in more detail on our project page and CVPR 2015 paper. The algorithms are described in more detail in the supplement. See below for install instructions.

If you use DP-vMF-means or DDP-vMF-means please cite:

Julian Straub, Trevor Campbell, Jonathan P. How, John W. Fisher III. 
"Small-Variance Nonparametric Clustering on the Hypersphere", In CVPR,
2015.

Dependencies

This code is dependent on Eigen3, Boost, CUDA, OpenCV and PCL. It has been tested on Ubuntu 14.04 with

• Eigen3 (3.0.5)
• Boost (1.54)
• CUDA (6.5)
• OpenCV (2.4)
• PCL (1.7)

Install

This package uses the pods build system. Used widely at CSAIL MIT the build system makes it easy to break up software projects into small packages that can be checked out and compiled automatically (see below).

• Linux:

  Install Eigen3, Boost, OpenCV, and PCL

  sudo apt-get install libeigen3-dev libboost-dev libopencv-dev libpcl-1.7-all-dev
  

  Install the appropriate CUDA version matching with your nvidia drivers. On our machines we use nvidia-340-dev with libcuda1-340 cuda-6-5 cuda-toolkit-6-5

  Clone this repository and compile the code:

  git clone git@github.com:jstraub/rtDDPvMF; cd rtDDPvMF;
  make checkout; make configure; make -j6; make install;
  

  Note that this will checkout several other necessary repositories. To update all repositories run

  make update; make configure; make -j6; make install;
  

Getting Started

Plug in your Kinect and run the following from the rtDDPvMF folder:

./build/bin/realtimeDDPvMF_openni --lambdaDeg 100 

Usage

./build/bin/realtimeDDPvMF_openni -h
Allowed options:
  -h [ --help ]               produce help message
  -K [ --K ] arg              K for spkm clustering
  -l [ --lambdaDeg ] arg      lambda in degree for DP-vMF-means and 
                              DDP-vMF-means
  -b [ --beta ] arg           beta parameter of the DDP-vMF-means
  -s [ --nFramesSurvive ] arg number of frames a cluster survives without 
                              observation
  -o [ --out ] arg            path to output file
  -d [ --display ]            display results
  -B [ --B ] arg              filter windows size B for guided filter
  --eps arg                   epsilon parameter for guided filter
  -f [ --f_d ] arg            focal length of depth camera