ccvw.2013.0012

A Novel Georeferenced Dataset for Stereo Visual Odometry

Ivan Krešo, Marko Ševrović and Siniša Šegvić

Abstract

In this work, we present a novel dataset for assessing the accuracy of stereo visual odometry. The dataset has been acquired by a small-baseline stereo rig mounted on the top of a moving car. The groundtruth is supplied by a consumer grade GPS device without IMU. Synchronization and alignment between GPS readings and stereo frames are recovered after the acquisition. We show that the attained groundtruth accuracy allows to draw useful conclusions in practice. The presented experiments address influence of camera calibration, baseline distance and zero-disparity features to the achieved reconstruction performance.

Files

Full Paper as PDF

BibTeX Citation

DOI

10.20532/ccvw.2013.0012

https://doi.org/10.20532/ccvw.2013.0012

BibTeX

@InProceedings{10.20532/ccvw.2013.0012,
  author =       {Ivan Kre{\v s}o and Marko {\v S}evrovi{\' c} and
                  Sini{\v s}a {\v S}egvi{\' c}},
  title =        {A Novel Georeferenced Dataset for Stereo Visual
                  Odometry},
  booktitle =    {Proceedings of the Croatian Compter Vision Workshop,
                  Year 1},
  pages =        {43-48},
  year =         2013,
  editor =       {Lon{\v c}ari{\' c}, Sven and {\v S}egvi{\' c},
                  Sini{\v s}a},
  address =      {Zagreb},
  month =        {September},
  organization = {Center of Excellence for Computer Vision},
  publisher =    {University of Zagreb},
  abstract =     {In this work, we present a novel dataset for
                  assessing the accuracy of stereo visual
                  odometry. The dataset has been acquired by a
                  small-baseline stereo rig mounted on the top of a
                  moving car. The groundtruth is supplied by a
                  consumer grade GPS device without
                  IMU. Synchronization and alignment between GPS
                  readings and stereo frames are recovered after the
                  acquisition. We show that the attained groundtruth
                  accuracy allows to draw useful conclusions in
                  practice. The presented experiments address
                  influence of camera calibration, baseline distance
                  and zero-disparity features to the achieved
                  reconstruction performance.},
  doi =          {10.20532/ccvw.2013.0012},
  url =          {https://doi.org/10.20532/ccvw.2013.0012}
}