New efficient solution to the absolute pose problem for camera with unknown focal length and radial distortion

Authors:
Martin Bujnak;Zuzana Kukelova;Tomas Pajdla
Affiliations:
-;Center for Machine Perception, Czech Technical University in Prague;Center for Machine Perception, Czech Technical University in Prague
Venue:
ACCV'10 Proceedings of the 10th Asian conference on Computer vision - Volume Part I
Year:
2010

Citing 9
Cited 1

Linear N-Point Camera Pose Determination

IEEE Transactions on Pattern Analysis and Machine Intelligence
Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography

Communications of the ACM
A New Efficient and Direct Solution for Pose Estimation Using Quadrangular Targets: Algorithm and Evaluation

IEEE Transactions on Pattern Analysis and Machine Intelligence
A complete symbolic-numeric linear method for camera pose determination

ISSAC '03 Proceedings of the 2003 international symposium on Symbolic and algebraic computation
Multiple View Geometry in Computer Vision

Multiple View Geometry in Computer Vision
PnP Problem Revisited

Journal of Mathematical Imaging and Vision
Photo tourism: exploring photo collections in 3D

ACM SIGGRAPH 2006 Papers
Automatic Generator of Minimal Problem Solvers

ECCV '08 Proceedings of the 10th European Conference on Computer Vision: Part III
Fast and Stable Polynomial Equation Solving and Its Application to Computer Vision

International Journal of Computer Vision

A Novel Solution to the P4P Problem for an Uncalibrated Camera

Journal of Mathematical Imaging and Vision

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we present a new efficient solution to the absolute pose problem for a camera with unknown focal length and radial distortion from four 2D-to-3D point correspondences. We propose to solve the problem separately for non-planar and for planar scenes. By decomposing the problem into these two situations we obtain simpler and more efficient solver than the previously known general solver. We demonstrate in synthetic and real experiments significant speedup as our new solvers are about 40× (non-planar) and 160× (planar) faster than the general solver. Moreover, we show that our two solvers can be joined into a new general solver, which gives comparable or better results than the existing general solver for of most planar as well as non-planar scenes.