Wyniki wyszukiwania - BazTech

1

A variantional approach to 3D line orientation estimation from motion

Yang L., Sahli H., Hao D. N.

Machine Graphics and Vision

|

2005

|

Vol. 14, No. 4

441-453

EN

A variational approach to estimating 3D line orientation from motion is presented. A 2D motion constraint on 3D lines regularized by a quadratic term is used to set up an objective functional. From its associated Euler-Lagrange equations, we develop a vector-valued diffusion model, with a reaction term based on the 2D motion constraint. Three separate diffusion processes, corresponding to each component of the 3D line orientation, are coupled with each other through the reaction term and evolve simultaneously. Each 3D line orientation is estimated separately. The regularization parameter is estimated by an L-curve, which provides a better estimation. Experimental results from image sequences indicate stability and accuracy of the approach.

2

Eudem2: the European Union in humanitarian demining. State-of-the-art on humanitarian demining technologies, products, services and practices in Europe

De Bruyn K., Barais M., Sahli H., Bruschini C., Wtorek J.

Zeszyty Naukowe Wydziału ETI Politechniki Gdańskiej. Technologie Informacyjne

|

2003

|

nr 2

597--604

3

Integration for 3D structure / motion estimation based on line drawings

Yang L., Sahli H.

Machine Graphics and Vision

|

2000

|

Vol. 9, No. 1/2

251-261

EN

A general framework for 3D structure / motion analysis is proposed upon integration of different visual modules. Line drawings are analyzed from motion point of view, and provide an effective means for 3D reconstruction. 2D motion based and feature correspondences based approaches are efficiently integrated with the line drawing interpretations under our framework. Normal flow of line segment is employed here to characterize 2D motion and the finite difference method is used for its estimation in a simple and practical way. To implement the proposed integration model, an incremential scheme is developed ti estimate 3D structure / motion from a sequence of images. Experiments on scenes containing polyhedral objects demonstrate the feasibility of the proposed scheme, and show that the integration of different visual modules gives better 3D structure / motion estrimations.