@inproceedings{sar-cat-min-sto-17-odo-rev,
  author = {Rafael Felipe V. Saracchini and Carlos A. Catalina and Rodrigo Minetto and Jorge Stolfi},
  title = {Real-Time Visual Odometry by Patch Tracking Using {GPU}-Based Perspective Calibration},
  booktitle = {VISIGRAPP Revised Selected Papers},
  year = 2017,
  month = aug,
  series = {Communications in Computer and Information Science},
  volume = {693},
  pages = {475--492},
  doi = {10.1007/978-3-319-64870-5_23},
  altkeys = {sar-cat-min-sto-16-odo-rev},
  abstract = {In this paper we describe VOPT (Visual Odometry by Patch Tracking), a robust algorithm for visual odometry, which is able to operate with sparse or dense maps computed by simultaneous localization and mapping (SLAM) algorithms. By using an iterative multi-scale procedure, VOPT is able to estimate the individual motion, photometric correction and reliability tracking confidence of a set of planar patches. In order to overcome the high computational cost of the patch adjustment, we use a GPU-based least-square solver, achieving real-time performance. The algorithm can also be used as a building block to other procedures for automatic initialization and recovery of 3D scene. Our tests show that VOPT outperforms the well-known PTAMM and the state-of-art ORB-SLAM algorithm in challenging videos using the same input maps.}
}