Model Tracking for Video-based Augmented Reality

A. Valinetti, A. Fusiello,and V. Murino

Dipartimento Scientifico e Tecnologico Università di Verona - Italy

Overview

This paper presents a technique for tracking complex objects (both polyhedral and smooth boundaries) in a monocular sequence. Our aim is to
use this model tracking method in an Augmented Reality context to compute the pose of a real object to be able to register it
with a synthetic one. A scalar score function for an object pose is defined, based on the local image gradient along the projected model
boundaries. A local search is then carried out in the configuration space of the pose to maximize the score. This technique is robust to
occlusions, since the whole object contour is used, not just few control points. The proposed method is effective yet simple. No image feature
extraction is necessary and no complex temporal evolution is used. Experimental results with real sequences shows the good performances of
our technique.

Method

A scalar evaluation score for an object pose is defined, based on the local image gradient along the projected model lines.
A local search is then carried out in the configuration space of the pose to maximize the score. In summary the features of the algorithm are:

no image feature extraction is necessary;
no complex temporal evolution like Kalman filter is used;
it copes with zoom cameras;
it copes with any type of object that OpenGL can render (polyhedra and smooth objects);
it is robust to occlusions, since use the whole object contour, not just few control points.

Results

sequence (MPEG)	model overlay	synthetic teapot overlay	note
Piramid_pan	X	X	camera is panning
Piramid_zoom	x	X	camera is zooming
Checkerboard	X	X	the checkerboard moves randomly
Pipes 1	X	not available	for smooth objects the silhouette is used instead of the wireframe
Pipes 2	X	not available	for smooth objects the silhouette is used instead of the wireframe

Reference paper

Valinetti, A. Fusiello, and V. Murino. Model tracking for video-based virtual reality. In Proceedings of the 11th International Conference on Image Analysis and Processing, pages 372-377, Palermo, Italy, 26-28 September 2001. IAPR, IEEE Computer Society Press. (PostScript) .