ParallaxBA

ParallaxBA: Bundle Adjustment using Parallax Angle Feature Parametrization.

Authors
Liang Zhao; Shoudong Huang; Yanbiao Sun; Gamini Dissanayake;

Get the Source Code via SVN
svn co https://svn.openslam.org/data/svn/ParallaxBA

Long Description
ParallaxBA is a new bundle adjustment algorithm based on parallax angle feature parametrization. It is demonstrated that the new parametrization can reliably represent both nearby and distant features under different camera motions. Because good information of the proposed feature parameters is always available, the singularity of the information matrix due to near zero parallax can be avoided. Moreover, using the new parameters results in better properties of the objective function as compared with those using traditional Euclidean XYZ and inverse depth parametrization. Thus ParallaxBA has better convergence properties as compared with existing BA algorithms.

Example Images

Parallax angle parametrization for feature

Result of Venice Dataset

Result of aerial photogrammetric 468 images dataset

Input Data
The input of ParallaxBA is the matched image points, camera calibration matrix, as well as the initial values of camera poses and features (optional).

Logfile Format
The format is similar to SBA.

Type of Map
3D point features and camera poses

Hardware/Software Requirements
Windows, Microsoft Visual Studio, or
Linux/Unix

Documentation
Document for ParallaxBA C/C++ Source Code

Papers Describing the Approach
Liang Zhao, Shoudong Huang, Lei Yan and Gamini Dissanayake: Parallax Angle Parametrization for Monocular SLAM, In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), pp. 3117-3124, 2011

Liang Zhao, Shoudong Huang, Yanbiao Sun, Lei Yan and Gamini Dissanayake: ParallaxBA: Bundle Adjustment using Parallax Angle Feature Parametrization, International Journal of Robotics Research (IJRR), 2015 (link)

License Information
This software is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
The authors allow the users of OpenSLAM.org to use and modify the source code for their own research. Any commercial application, redistribution, etc has to be arranged between users and authors individually and is not covered by OpenSLAM.org.

ParallaxBA by Liang Zhao, Shoudong Huang, Yanbiao Sun, Gamini Dissanayake is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.


*** OpenSLAM.org is not responsible for the content of this webpage ***
*** Copyright and V.i.S.d.P.: Liang Zhao; Shoudong Huang; Yanbiao Sun; Gamini Dissanayake; ***