This project addresses the problem of automatically generating high quality class independent object bounding boxes and segmentations using color and depth images of indoor scenes. The software is licensed under the GNU General Public License. If you use this project for your research, please cite:
@article{deng2016unsupervised,
title={Unsupervised object region proposals for RGB-D indoor scenes},
author={Deng, Zhuo and Todorovic, Sinisa and Latecki, Longin Jan},
journal={Computer Vision and Image Understanding},
year={2016},
publisher={Elsevier}
}1 Get the NYU data:
wget http://www.cis.temple.edu/~latecki/TestData/NYUv2data.zip2 Get precomputed results (plane segmentations, bounding boxes, object segments):
wget http://www.cis.temple.edu/~latecki/TestData/NYUv2result.zip