Multiple ground plane estimation for 3D scene understanding using a monocular camera

Authors:
Sangdoo Yun;Soo Wan Kim;Kwang Moo Yi;Haan-ju Yoo;Jin Young Choi
Affiliations:
Seoul National University, Seoul, Korea;Seoul National University, Seoul, Korea;Seoul National University, Seoul, Korea;Seoul National University, Seoul, Korea;Seoul National University, Seoul, Korea
Venue:
Proceedings of the 27th Conference on Image and Vision Computing New Zealand
Year:
2012

Citing 7
Cited 0

Mean Shift, Mode Seeking, and Clustering

IEEE Transactions on Pattern Analysis and Machine Intelligence
Multiple View Geometry in Computer Vision

Multiple View Geometry in Computer Vision
Bayesian Autocalibration for Surveillance

ICCV '05 Proceedings of the Tenth IEEE International Conference on Computer Vision - Volume 2
A Dynamic Bayesian Network Model for Autonomous 3D Reconstruction from a Single Indoor Image

CVPR '06 Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 2
Robust Auto-Calibration from Pedestrians

AVSS '06 Proceedings of the IEEE International Conference on Video and Signal Based Surveillance
Learning long-range vision for autonomous off-road driving

Journal of Field Robotics - Special Issue on LAGR Program, Part II
Make3D: Learning 3D Scene Structure from a Single Still Image

IEEE Transactions on Pattern Analysis and Machine Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we propose a method for estimation of multiple ground planes using a stationary monocular camera. To estimate multiple ground planes, we perform three major steps. First, to estimate the number of ground planes, we create a histogram of votes with vanishing points and perform mean-shift clustering on this histogram. Second, to recover the active regions of multiple ground planes, we perform back-projection with the votes from the first step to extract trajectories which support each ground plane. We then estimate the active regions of each ground planes with these supporting trajectories. Finally, we efficiently normalize the relative depths of multiple ground planes with the speed of moving objects in the ground planes. In the experiments, we demonstrate that our method successfully estimates multiple ground planes and their relative depths.