Abstract:For the existing monocular vision methods of pose (or position and orientation) measurement based on the Perspective-n-Point problem, camera calibrations and image corrections are complicated and difficult due to the lens distortion. Additionally, as their theoretical models, the pinhole imaging model and the collinear equation only offer an ideal approximation for the imaging process of the area array camera. Thus, the measurement possesses inevitable systematic errors. To avoid these problems, a new pose measuring method is proposed, which replaces the area array camera and its collinear equation with two line array cameras and their incident light plane equations, respectively. For these two line array cameras, their mechanical assembly requirements and calibration steps are also analysed; they only require that the two optical centre lines of their cylindrical lenses be perpendicular to each other, and only need to be calibrated for the relationship between the coordinate of the image line and the incident angle of the incident light plane. The experiment results of P4P pose measurement with the proposed method showed only significant random errors and no significant systematic error, confirming that this method has eliminated the problems caused by the image distortion and model approximation. (C) 2013 Society of Photo-Optical Instrumentation Engineers (SPIE). [DOI: 10.1117/1.OE.52.1.013604]

Main developments of perspective-n-point problems

Finding All the Solutions of PnP Problem

ScPnP: A Non-Iterative Scale Compensation Solution for PnP Problems

A Robust and Fast Method to the Perspective-n-Point Problem for Camera Pose Estimation

The Perspective-3-Point Problem When Using a Planar Mirror

Stereo Calibration and Rectification for Omnidirectional Multi-Camera Systems

A robust direct linear transformation for camera pose estimation using points

MLPnP - A Real-Time Maximum Likelihood Solution to the Perspective-n-Point Problem

DeepPanoContext: Panoramic 3D Scene Understanding with Holistic Scene Context Graph and Relation-based Optimization

Generalized Maximum Likelihood Estimation for Perspective-n-Point Problem

CPnP: Consistent Pose Estimator for Perspective-n-Point Problem with Bias Elimination

Optimal DLT-based Solutions for the Perspective-n-Point

Problem-solving of PnP based on an improved DLT algorithm

Z-shaped Perspective-Three-Line Problem's Unique Solution Conditions

A Special Unique Solution Case of the Perspective-Three-Point Problem for External Parameter Calibration of an Omnidirectional Camera

PLMP -- Point-Line Minimal Problems in Complete Multi-View Visibility

Perspective-N-Point Pose Measurement with Two Line Array Cameras

PL${}_{1}$P -- Point-line Minimal Problems under Partial Visibility in Three Views

PL P: Point-Line Minimal Problems under Partial Visibility in Three Views

Perspectives Of Non-Photorealistic Computer Graphics

Learning 2D-3D Correspondences To Solve The Blind Perspective-n-Point Problem