2D to 3D Line-Based Registration with Unknown Associations via Mixed-Integer Programming
Steven A. Parkison, Jeffrey M. Walls, Ryan W. Wolcott, Mohammad Saad, Ryan M. Eustice
Abstract
Determining the rigid-body transformation be-tween 2D image data and 3D point cloud data has applications for mobile robotics including sensor calibration and localizing into a prior map. Common approaches to 2D-3D registration use least-squares solvers assuming known associations often provided by heuristic front-ends, or iterative nearest-neighbor. We present a linear line-based 2D-3D registration algorithm formulated as a mixed-integer program to simultaneously solve for the correct transformation and data association. Our formulation is explicitly formulated to handle outliers, by modeling associations as integer variables. Additionally, we can constrain the registration to SE(2) to improve runtime and accuracy. We evaluate this search over multiple real-world data sets demonstrating adaptability to scene variation.
BibTeX
@inproceedings{icra2020_2dto3dlinebasedr,
title = {2D to 3D Line-Based Registration with Unknown Associations via Mixed-Integer Programming},
author = {Steven A. Parkison and Jeffrey M. Walls and Ryan W. Wolcott and Mohammad Saad and Ryan M. Eustice},
booktitle = {ICRA 2020},
year = {2020}
}