Readers should be aware of some of the issues raised by the electronic dissemination of published papers. In particular, readers should know that:
The documents contained in these directories are included by the contributing authors as a means to ensure timely dissemination of scholarly and technical work on a non-commercial basis. Copyright and all rights therein are maintained by the authors and by other copyright holders, notwithstanding that they have offered their works here electronically. It is understood that all persons copying this information will adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.

Publications

Homography from a Vanishing Point in Urban Scenes
IEEE RSJ/International conference on Intelligent Robot and System (IROS'03)
Abstract: In this paper, we address the problem of computing the ego-motion of a vehicle in an urban environment using dynamic vision. We assume a planar piecewise world where the planes are mainly distributed along three principal directions corresponding to the axes of a reference frame linked to the ground plane with a vertical z-axis. We aim to estimate both the motion of the car and the principal planes in the scene corresponding to the road and the frontages of the building from a sequence of images provided by an on-board uncalibrated camera. In this paper, we present preliminary results concerning the robust segmentation of the road using projective properties of the scene. We develop a two-stage algorithm in order to increase robustness. The first stage detects the borders of the road using a contour-based approach and primarily allows us to estimate the Dominant Vanishing Point (DVP). The DVP and the borders of the road are then used to constrain the region where the points of interest, corresponding to the road lane markers, can be extracted. The second stage uses a robust technique based on projective invariant to match the lines and points between two consecutive images in the sequence. Finally, we compute the homography relating the points and lines lying on the road into the two images.

document (700 kb), presentation (3.5 Mb),

Détection robuste du plan de la route en milieu urbain
Congrès Francophone de Reconnaissance de Formes et Intelligence Artificielle (RFIA'04)
Résumé : Cet article décrit un algorithme de segmentation du plan de la route dans des images de scènes urbaines. Nous supposons que l'environnement contient une majorité de contours alignés selon les 3 directions principales. L'algorithme se décompose en deux étapes. La première détecte conjointement les lignes de fuite qui convergent vers le point de fuite dominant. Les lignes de fuite sont issues d'une détection de contours à laquelle nous appliquons des contraintes spécifiques aux scènes urbaines. Nous recherchons alors des points d'intérêts dans la région de l'image délimitée par les lignes de fuite coplanaires. Lors de la deuxième étape, nous mettons en correspondance l'ensemble des caractéristiques détectées (points d'intérêts et lignes de fuite) en utilisant les propriétés d'invariance projective entre deux prises de vues d'une même scène de façon à estimer l'homographie induite par le plan de la route.

document (3,4 Mb), presentation (3.5 Mb),

Trajectography of an Uncalibrated Stereo Rig in Urban Environments
IEEE RSJ/International conference on Intelligent Robot and System (IROS'04)
Abstract: This paper describes an original method to compute the relative motion of an uncalibrated stereo rig in urban environments from features lying on the road. The extraction of significant reliable features on the road remains the critical step of this method. We nevertheless detect them according to the stereo constraints and an a priori knowledge on the scene. The motion between two frames of the stereo rig is considered as rigid: the homography computation is enforced by the redundancy of the feature locations in multiple views. The method has been tested on video sequences recorded from a test vehicle that was driven in an urban environments. Promising results from these experiments will be presented.

document (404 kb), presentation (1.4 Mb),

Localisation robuste d'un véhicule en environnement urbain à partir d'un système de stéréo-vision
PhD thesis prepared in the ICARE project-team, INRIA Sophia-Antipolis, july 2005

document (20.7 Mb), presentation (6.9 Mb),

Reconstruction of the road plane with an embedded stereo-rig in urban environments
IEEE Intelligent Vehicles Symposium (IV'06)
Award of the best interactive session paper
Abstract: We present in this article an original method to reconstruct the road in the specific context of urban environment thanks to the data provided by an uncalibrated stereo-vision system. The method consists on extracting then tracking features (points, lines) from the road and estimate the homography induced by the plane between two poses. The purposed method copes with the dense traffic conditions: the free space required (first ten meters in front of the vehicle) is slightly equivalent to the security distance between two vehicles. Experimental issues from real data are presented and discussed.

document (1.4 Mb), poster (2.4 Mb),

Vehicle Trajectory from an Uncalibrated Stereo-Rig with Super-Homography
IEEE RSJ/International conference on Intelligent Robot and System (IROS'06)
Abstract: A vision-based method to estimate the vehicle displacement in urban environments is decribed when no GPS data are available. The method consists on tracking coplanar features of the static environment to estimate the homography induced by the tracked plane between two views. The road is the common link between all the urban scenes therefore we mainly focus on it to develop the method. When cameras have large fields of view, the method can be used to track other planes like those extracted from the frontages. The method can be assumed as a reliable visual odometer when no data from GPS system are available and is particularly well-adapted to the urban traffic: the homography computation requires only a few meters of free space in front of the vehicle.

document (872 kb), presentation (12.4 Mb),

Obstacle Detection from IPM and Super-Homography
IEEE RSJ/International conference on Intelligent Robot and System (IROS'07)
Abstract: We present in this article a simple method toestimate an IPM view from an embedded camera. The method is based on the tracking of the road markers assuming that the road is locally planar. Our aim is the development of a free-space estimator which can be implemented in an Autonomous Guided Vehicle to allow a safe path planning. Opposite to most of the obstacle detection methods which make assumptions on the shape or height of the obstacles, all the scene elements above the road plane (particularly kerbs and poles) have to be detected as obstacles. Combined with the IPM tranformation, this obstacle detection stage can be viewed as the first stage of a free-space estimator dedicated to AGV in the complex urban environments.

document (1.1 Mb), presentation (8 Mb),

Talk

What Can Be Done with an Embedded Stereo-Rig in Urban Environments ?
School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore, 20th june 2006

presentation (14.7 Mb)