Academic
Publications
Localization of mobile robots with omnidirectional vision using Particle Filter and iterative SIFT

Localization of mobile robots with omnidirectional vision using Particle Filter and iterative SIFT,10.1016/j.robot.2006.04.018,Robotics and Autonomous

Localization of mobile robots with omnidirectional vision using Particle Filter and iterative SIFT   (Citations: 24)
BibTex | RIS | RefWorks Download
ABSTRACT The Scale Invariant Feature Transform, SIFT, has been success- fully applied to robot localization. Still, the number of features extracted with this approach is immense, especially when dealing with omnidirectional vision. In this work, we propose a new ap- proach that reduces the number,of features generated by SIFT as well as their extraction and matching,time. With the help of a par- ticle filter, we demonstrate that we can still localize the mobile robot accurately with a lower number,of features.
Journal: Robotics and Autonomous Systems - RaS , vol. 54, no. 9, pp. 758-765, 2006
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
    • ...Also known as the bootstrapping filter, survival-of-thefittest, and the CONDENSATION algorithm, the particle filter andits variations havebeen appliedto a wide variety of applications [4], including finance theory, audio recognition, fault detection, and robot localization [5], [6], [7]...

    James Anthony Brownet al. A Framework for 3D Model-Based Visual Tracking Using a GPU-Accelerated...

    • ...The SIFT algorithm usually produces a redundant number of features that impose high computational cost in the other stages of feature description, matching, and mismatch elimination [36]‐[38]...
    • ...The quality control of standard SIFT features is limited to the contrast and ratio between the principal curvatures (Tc and Tr thresholds), which usually leads to a high redundancy of weak points that are prone to fail in the correspondence module [37], [38]...
    • ...Also, Tamimi et al. [38] proposed an iterative SIFT algorithm for localization of mobile robots that reduces the number of features generated by the SIFT algorithm as well as their extraction and matching time...

    Amin Sedaghatet al. Uniform Robust Scale-Invariant Feature Matching for Optical Remote Sen...

    • ...Modified SIFT by Andreasson et al. (2005), Iterative SIFT by Tamimi et al. (2005)) have been proposed to reduce the computational efforts of the feature extraction and matching process, and applied in almost real time robot localization...
    • ...Tamimi et al. (2005) built up a collection of histograms of descriptors based on SIFT interest points and their local gradient orientations...
    • ...Tamimi et al. 2005 and Leprêtre et al. 2000) when an image signature is built based on a collection of image local characteristics; or as global methods (e.g...
    • ...Tamimi et al. (2005) where the invariant signature is built up a collection of histograms of descriptors relating the local structure with reference to local gradi-...

    Ouiddad Labbani-Igbidaet al. Haar invariant signatures and spatial recognition using omnidirectiona...

    • ...In the second category, more sophisticated features are used [13, 14, 15, 7]. In [13], Se et al. propose a vision-based simultaneous localization and mapping (SLAM) system by tracking the SIFT features...
    • ...Tamimi et al. [14] propose an approach that reduces the number of features generated by SIFT, and with the help of a particle filter, the robot location can still be estimated accurately...

    Wei Guanet al. Computationally efficient retrieval-based tracking system and augmente...

    • ...Over the past years, researchers have developed many such algorithms such as [2, 5, 6, 12, 13, 16]...

    Wei Guanet al. Recognition-driven 3D navigation in large-scale virtual environments

Sort by: