We consider the problem of detecting and accounting for the presence of occluders in a 3D scene based on silhouette cues in video streams obtained from multiple, calibrated views....
In this work, we present an approach to jointly segment a rigid object in a 2D image and estimate its 3D pose, using the knowledge of a 3D model. We naturally couple the two proces...
Samuel Dambreville, Romeil Sandhu, Anthony J. Yezz...
— A points surface reconstruction algorithm of 3D object models from multiple silhouettes is proposed in this paper. Some images of the target object are taken from a circular tr...
Abstract. In this work, we present an approach to jointly segment a rigid object in a two-dimensional (2D) image and estimate its three-dimensional (3D) pose, using the knowledge o...
Samuel Dambreville, Romeil Sandhu, Anthony J. Yezz...
A model-based approach to reconstruction of 3D human arm motion from a monocular image sequence taken under orthographic projection is presented. The reconstruction is divided into...