Disparity Sliding Window: Object Proposals from Disparity Images

This respository provides code for our disparity sliding window method, a real-time proposal generator. Paper is submitted for IROS 2018. Code is written in C++ and also provides a wrapper for usage in Python.

Contents

1. Requirements:
2. Installation:
3. Demo:
4. Citation:

Requirements:

1. OpenCV (we used 2.4)
2. Boost

Installation:

1. Clone the DSW respository

git clone https://github.com/julimueller/disparity-sliding-window

2. Build everything

2. mkdir build && cd build
3. cmake .. -DCMAKE_INSTALL_PREFIX="YOUR_PATH" && make -j12 install

Note: "YOUR_PATH" has to be in LD_LIBRARY_PATH.

Demo:

Demo in C++:

disparity_sliding_window_test

Demo in Python:

PYTHONPATH=PYTHONPATH:"dir_to_dsw_python.so"
cd disparity-sliding-window/python
python dsw_test.py

Results should be close to the following:

Citation:

Of course we would be happy to be cited in your work. Use the following BibTeX entry:

@article{DBLP:journals/corr/abs-1805-06830,
  author    = {Julian M{\"{u}}ller and
               Andreas Fregin and
               Klaus Dietmayer},
  title     = {Disparity Sliding Window: Object Proposals From Disparity Images},
  journal   = {CoRR},
  volume    = {abs/1805.06830},
  year      = {2018},
  url       = {http://arxiv.org/abs/1805.06830},
  archivePrefix = {arXiv},
  eprint    = {1805.06830},
  timestamp = {Tue, 05 Jun 2018 18:50:11 +0200},
  biburl    = {https://dblp.org/rec/bib/journals/corr/abs-1805-06830},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}