Fakhreddine Ababsa

Robust line tracking using a particle filter for camera pose estimation

This paper presents a robust line tracking approach for camera pose estimation which is based on particle filtering framework. Particle filters are sequential Monte Carlo methods based on point mass (or particle) representations of probability densities, which can be applied to any state-space model. Their ability to deal with non-linearities and non-Gaussian statistics allows to improve robustness comparing to existing approaches, such as those based on the Kalman filter. We propose to use the particle filter to compute the posterior density for the camera 3D motion parameters. The experimental results indicate the effectiveness of our approach and demonstrate its robustness even when dealing with severe occlusion.

Robust Extended Kalman Filtering for camera pose tracking using 2D to 3D lines correspondences

In this paper we present a new robust camera pose estimation approach based on 3D lines tracking. We used an Extended Kalman Filter (EKF) to incrementally update the camera pose in real-time. The principal contributions of our method includes first, the expansion of the RANSAC scheme in order to achieve a robust matching algorithm that associates 2D edges from the image with the 3D line segments from the input model. And second, a new framework for camera pose estimation using 2D-3D straight-lines within an EKF. Experimental results on real image sequences are presented to evaluate the performances and the feasibility of the proposed approach.

Robust camera pose tracking for augmented reality using particle filtering framework

In this paper, we present new solutions for the problem of estimating the camera pose using particle filtering framework. The proposed approach is suitable for real-time augmented reality (AR) applications in which the camera is held by the user. This work demonstrates that particle filtering improve the robustness of the tracking comparing to existing approaches, such as those based on the Kalman filter. We propose a tracking framework for both points and lines features, the particle filter is used to compute the posterior density for the camera 3D motion parameters. We also analyze the sensitivity of our technique when outliers are present in the match data. Outliers arise frequently due to incorrect correspondences which occur because of either image noise or occlusion. Results from real data in an augmented reality setup are then presented, demonstrating the efficiency and robustness of the proposed method.

Vision-Based Tracking for Mobile Augmented Reality

Augmented Reality Systems (ARS) attempt to enhance humans’ perception of their indoors and outdoors working and living environments and understanding of tasks that they need to carry out. The enhancement is effected by complementing the human senses with virtual input. For example, when the human visual sense is enhanced, an ARS allows virtual objects to be superimposed on a real world by projecting the virtual objects onto real objects. This provides the human user of the ARS with additional information that he/she could not perceive with his/her senses. In order to receive the virtual input and sense the world around them augmented with real time computer-generated features, users of an ARS need to wear special equipment, such as head-mounted devices or wearable computing gears. Tracking technologies are very important in an ARS and, in fact, constitute one challenging research and development topic. Tracking technologies involve both hardware and software issues, but in this chapter we focus on tracking computation. Tracking computation refers to the problem of estimating the position and orientation of the ARS user’s viewpoint, assuming the user to carry a wearable camera. Tracking computation is crucial in order to display the composed images properly and maintain correct registration of real and virtual worlds. This tracking problem has recently become a highly active area of research in ARS. Indeed, in recent years, several approaches to vision-based tracking using a wearable camera have been proposed, that can be classified into two main categories, namely “marker-based tracking” and “marker-less tracking.” In this chapter, we provide a concise introduction to vision-based tracking for mobile ARS and present an overview of the most popular approaches recently developed in this research area. We also present several practical examples illustrating how to conceive and to evaluate such systems.

Inertial and vision head tracker sensor fusion using a particle filter for augmented reality systems

A basic problem with augmented reality systems using head-mounted displays (HMDs) is the perceived latency or lag. This delay corresponds to the elapsed time between the moment when the users head moves and the moment of displaying the corresponding virtual objects in the HMD. One way to eliminate or reduce the effects of system delays is to predict the future head locations. Actually, the most used filter to predict head motion is the extended Kalman filter (EKF). However, when dealing with nonlinear models (like head motion) in state equation and measurement relation and a non Gaussian noise assumption, the EKF method may lead to a non optimal solution. In this paper, we propose to use sequential Monte Carlo methods, also known as particle filters to predict head motion. These methods provide general solutions to many problems with any nonlinearities or distributions. Our purpose is to compare, both in simulation and in real task, the results obtained by particle filter with those given by EKF.

Outdoor augmented reality system for geological applications

Augmented reality has been shown to be useful in many application areas such as maintenance skills, urban planning, interior design and entertainment, etc. The development of the mobile augmented reality in these last years is due to the evolution of the technology at various levels: from sensors (GPS, inertial sensor, etc.) to mobile devices (Tablet-PC, PDA, etc.). In this paper, we present a mobile augmented reality system dedicated to outdoor applications. It encompasses a localization process based on an assistance scheme and combining several data acquired from a hybrid sensor (camera, GPS and inertial sensor). The visualization and interaction are performed using a tablet-PC. This system has been tested in an application intended to geologists in order to monitor and supervise the restoration of a castle on a long term. This paper details various components of the system, some results that were obtained and how the whole application was evaluated through end-user.

Marker Less Vision-Based Tracking of Partially Known 3D Scenes for Outdoor Augmented Reality Applications

This paper presents a new robust and reliable marker less camera tracking system for outdoor augmented reality using only a mobile handheld camera. The proposed method is particularly efficient for partially known 3D scenes where only an incomplete 3D model of the outdoor environment is available. Indeed, the system combines an edge-based tracker with a sparse 3D reconstruction of the real-world environment to continually perform the camera tracking even if the model-based tracker fails. Experiments on real data were carried out and demonstrate the robustness of our approach to occlusions and scene changes.

Photometric aspects: a new approach for 3D free form object recognition using a single luminance image

3D free form object recognition is one of the most difficult problems in computer vision. In this paper we present a new approach which exploit only one luminance image of a complex object to recognize it in the scene by identifying its appearance in the input image. For that, we construct a photometric (non geometric) projective invariant to perform matching between local regions of the object in the image and those of the model. We propose an original method based on what we called photometric aspects to construct a discriminative data base of the 3D object model. We demonstrate the effectiveness of our approach while implementing it with complex free form objects and we present some obtained results.