0000000000976357
AUTHOR
David Fofi
Non conventional Imaging Systems for 3D Digitization of transparent and/or specular manufactured objects
International audience; 3D scanning has been investigated for several years and most of the proposed approaches assume a diffuse or near diffuse reflectance of the object's surface, also called cooperative surfaces. For the case of "non cooperative surfaces", such as transparent objects or mirror-like surfaces, usually, a thin layer of powder is sprayed onto the object surface (to make it opaque and diffuse) prior to its digitization. This extra step is troublesome, time consuming (the object needs to be cleaned afterwards), and the final accuracy is often dependent on the powder thickness and its homogeneousness. To avoid this step, various methods have been investigated over the last few …
Using Visual Saliency for Object Tracking with Particle Filters
International audience; This paper presents a robust tracking method based on the integration of visual saliency information into the particle filter framework. While particle filter has been successfully used for tracking non-rigid objects, it shows poor performances in the presence of large illumination variation, occlusions and when the target object and background have similar color distributions. We show that considering saliency information significantly improves the performance of particle filter based tracking. In particular, the proposed method is robust against occlusion and large illumination variation while requiring a reduced number of particles. Experimental results demonstrate th…
Short Base-line Line Matching for Central Imaging Systems
International audience
Numérisation 3D intelligente : deux algorithmes automatiques et progressifs
This paper presents two methods for the digitization of 3D objects without prior knowledge on their shape. The first method is based on the bounding box of the object during the acquisition steps. The second method is a combination of two concepts : the Mass Vector Chains (MVC) to define the orientation of the scanned part and the Mesurability Matrix to determine the coverage rate for each view point. Tests with simulated objects were performed and results of digitization are provided.
European Programs and their Extension in the field of Computer Vision, Color and Robotics
International audience; Over the last 4 years, the University of Burgundy (uB) has intensified its international oriented actions in the field of Computer Vision and Robotics through further development of the educational network, partnerships, programs and communication. The creation of European programs, such as the Erasmus Mundus program and the Thematic Networks (Life Long Learning) allowed us to develop new programs, building up valuable academic and project administration experience as well as developing new actions for incoming foreign students. In this paper we describe some of our activities and those of the University of Saint Etienne (UJM) and University of Genova (UNIGE), and th…
Line Matching across Catadioptric Images under Short Baseline Motion
International audience
Line Matching across Catadioptric Images under Short-Baseline Motion
International audience; Line matching across catadioptric images using line intersections with focus on short baseline motion of the imaging system is proposed. The relationship between images of constructing lines of an intersection on unitary sphere is studied and angle consistency and antipodal consistency are introduced as two useful properties which can be employed to find putative intersections correspondence and remove the outliers. The necessary equations for adapting boundary of a rectangular patch to the geometry of the catadioptric images are also derived and a standard correlation is used for measuring similarities of the intersections. Experimental results on both synthetic and…
Self-Calibration of PTZ Camera using New LMI Constraints
International audience
Control of a PTZ camera in a hybrid vision system
In this paper, we propose a new approach to steer a PTZ camera in the direction of a detected object visible from another fixed camera equipped with a fisheye lens. This heterogeneous association of two cameras having different characteristics is called a hybrid stereo-vision system. The presented method employs epipolar geometry in a smart way in order to reduce the range of search of the desired region of interest. Furthermore, we proposed a target recognition method designed to cope with the illumination problems, the distortion of the omnidirectional image and the inherent dissimilarity of resolution and color responses between both cameras. Experimental results with synthetic and real …
Image filtering in catadioptric plane
International audience
Current State-of-the-Art of Vision-based SLAM
International audience