Sylvain Brandel

STIGMA: a 4-dimensional modeller for animation

This article presents an animation method and a software based on 4-D object modelling. This method uses a topological model and a free-form deformation model. It defines a set of 4-D modelling operations for the construction of 4-D objects. The resulting animation is a sequence of 3-D objects, obtained by successive sections of 4-D object. Using 4-D objects allows topological modification of the animated 3-D objects, such as the merging and the splitting of volumes.

vrLib: A Designer Oriented Interaction and 3D User Interface Library

In this paper we propose vrLib, a toolkit which aims to make 3D graphical user interface and interaction very easy to design in virtual environments. This project provides well known and new interaction techniques and ready to use widgets, which are totally hardware independent. The main contribution allows to elaborate complex 3D user interfaces in same way it is done in 2D. This toolkit provides adaptive interaction, which means interaction techniques can automatically adapt their behavior to 2D and 3D widgets. This short paper describes the vrLib architecture for construction of 3D interactive applications.

Incremental Updating of 3D Topological Maps to Describe Videos

A topological map is an efficient mathematical model for representing an image subdivision where all cells and adjacency relations between elements are represented. It has been proved to be a very good tool for video processing when video is seen as a 3D image. However the construction of a topological map for representing a video needs the availability of the complete image sequence. In this paper we propose a procedure for online updating a topological map in order to build it as the video is produced, allowing to use it in real time.

Fast environment extraction for lighting and occlusion of virtual objects in real scenes

Augmented reality aims to insert virtual objects in real scenes. In order to obtain a coherent and realistic integration, these objects have to be relighted according to their positions and real light conditions. They also have to deal with occlusion by nearest parts of the real scene. To achieve this, we have to extract photometry and geometry from the real scene. In this paper, we adapt high dynamic range reconstruction and depth estimation methods to deal with real-time constraint and consumer devices. We present their limitations along with significant parameters influencing computing time and image quality. We tune these parameters to accelerate computation and evaluate their impact on the resulting quality. To fit with the augmented reality context, we propose a real-time extraction of these information from video streams, in a single pass.