3D Pose Identification of Micro- and Nanowires in Fluid Suspensions

AbstractIn the literature of computer vision, computer graphics and robotics, the use of quaternions is exclusively related to 3D rotation representation and interpolation. In this research we found how epipoles in multi-camera systems can be used to represent camera poses in the quaternion domain. The rotational quaternion is decomposed in two epipole rotational quaternions and one z axis rotational quaternion. Quadratic form of the essential matrix is also related to quaternion factors. Thus, five pose parameters are distributed into three independent rotational quaternions resulting in measurement error separation at camera pose identification and greater flexibility at virtual camera animation. The experimental results refer to the design of free viewpoint television.

Download Full-text

Low Density Feature Point Matching for Articulated Pose Identification

Procedings of the British Machine Vision Conference 2002 ◽

10.5244/c.16.66 ◽

2002 ◽

Cited By ~ 4

Author(s):

H. Holstein ◽

B. Li

Keyword(s):

Feature Point ◽

Low Density ◽

Point Matching ◽

Feature Point Matching ◽

Pose Identification

Download Full-text

Adaptive Trees and Pose Identification from External Contours of Polyhedra

Lecture Notes in Computer Science - Deep Structure, Singularities, and Computer Vision ◽

10.1007/11577812_14 ◽

2005 ◽

pp. 157-168 ◽

Cited By ~ 3

Author(s):

Yannick L. Kergosien

Keyword(s):

Pose Identification

Download Full-text

REAL-TIME FACIAL POSE IDENTIFICATION WITH HIERARCHICALLY STRUCTURED ML POSE CLASSIFIER

International Journal of Pattern Recognition and Artificial Intelligence ◽

10.1142/s0218001404003125 ◽

2004 ◽

Vol 18 (02) ◽

pp. 127-142 ◽

Cited By ~ 3

Author(s):

JAE-WON SUNG ◽

DAIJIN KIM

Keyword(s):

Real Time ◽

High Speed ◽

Tracking System ◽

Computation Time ◽

Feature Space ◽

Identification Problem ◽

Pose Tracking ◽

Time Simulation ◽

The Face ◽

Pose Identification

Since pose-varying face images form nonlinear convex manifold in high dimensional image space, it is difficult to model their pose distribution in terms of a simple probabilistic density function. To solve this difficulty, we divide the pose space into many constituent pose classes and treat the continuous pose estimation problem as a discrete pose-class identification problem. We propose to use a hierarchically structured ML (Maximum Likelihood) pose classifiers in the reduced feature space to decrease the computation time for pose identification, where pose space is divided into several pose groups and each group consists of a number of similar neighboring poses. We use the CONDENSATION algorithm to find a newly appearing face and track the face with a variety of poses in real-time. Simulation results show that our proposed pose identification using the hierarchically structured ML pose classifiers can perform a faster pose identification than conventional pose identification using the flat structured ML pose classifiers. A real-time facial pose tracking system is built with high speed hierarchically structured ML pose classifiers.

Download Full-text