In this paper we present several interest points detectors and we analyze their suitability when used as landmark extractors for vision-based simultaneous localization and mapping (vSLAM). For this purpose, we evaluate the detectors according to their repeatability under changes in viewpoint and scale. These are the desired requirements for visual landmarks. Several experiments were carried out...

We report on a stereo system for 3D detection and tracking of pedestrians in urban traffic scenes. The system is built around a probabilistic environment model which fuses evidence from dense 3D reconstruction and image-based pedestrian detection into a consistent interpretation of the observed scene, and a multi-hypothesis tracker to reconstruct the pedestrians’ trajectories in 3D coordinates ...

A new algorithm is presented for feature point based motion tracking in long image sequences. Dynamic scenes with multiple, independently moving objects are considered in which feature points may temporarily disappear, enter and leave the view field. The existing approaches to feature point tracking [6, 3, 5, 4] have limited capabilities in handling incomplete trajectories, especially when the ...

This contribution gives an overview of automatic 3D scene reconstruction and visualisation from uncalibrated and handheld camera image sequences. We address specifically the problems that are associated with calibration and visual-geometric reconstruction of complex scenes with occlusions and view-dependent surfaces. The scene is then represented by large sets of calibrated real viewpoints with...

This paper proposes novel metrics to evaluate the performance of object detection algorithms in video sequences. The proposed metrics allow to characterize the methods being used and classify the types of errors into region splitting, merging or merge-split, detection failures and false alarms. This methodology is applied to characterize the performance of five segmentation algorithms. These te...

This paper deals with stereo correspondence search, using graph cuts and belief propagation, for estimating depth maps. The results following different preprocessing steps are evaluated, using the quality of the disparity map. Running times are also investigated. For evaluation purposes, different kinds of images have been used: reference images from the Middlebury Stereo website, synthetic dri...

Statistical background modeling is a fundamental and important part for many visual tracking systems and other computer vision applications. This paper presents an effective and adaptive background modeling method for detecting foreground objects in both static and dynamic scenes. The proposed method computes SAmple CONsensus (SACON) of the background samples and estimates a statistical model p...

This paper presents a fully automatic image mosaicing method for needs of wide-area video surveillance. A pure featurebased approach was adopted for finding the registration between the images. This approach provides us with several advantages. Our method is robust against illumination variations, moving objects, image rotation, image scaling, imaging noise, and is relatively fast to calculate....

This paper proposes novel metrics to evaluate the performance of object detection algorithms in video sequences. The proposed metrics allow to characterize the methods being used and classify the types of errors into region splitting, merging or merge-split, detection failures and false alarms. This methodology is applied to characterize the performance of five segmentation algorithms. These te...

Estimating motion in scenes containing multiple motions remains a diicult problem for computer vision. Here we describe a novel recurrent network architecture which solves this problem by simultaneously estimating motion and segmenting the scene. The network is comprised of locally connected units which carry out simple calculations in parallel. We present simulation results illustrating the su...