Improving Data Association in Vision-based SLAM

Gil, Arturo; Reinoso, Óscar; Mozos, Óscar Martı́nez; Stachniss, Cyrill; Burgard, Wolfram

doi:10.1109/iros.2006.282483

Cited by 42 publications

(55 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Later, Se et al [24] used SIFT features as landmarks in the 3D space. Little et al [14] and Gil et al [7] additionally tracked the detected SIFT points to keep the most robust ones. Jensfelt et al [11] use a rotationally variant version of SIFT in combination with a Harris-Laplace detector for monocular SLAM.…”

Section: Related Workmentioning

confidence: 99%

“…The algorithm was initially presented by Lowe [15] and used in object recognition tasks. Lately, it has been used in visual SLAM applications [25,29,7]. In this work we separate the detection process from the description, thus when used as a detector, points are extracted using a DoG function.…”

Section: Siftmentioning

confidence: 99%

“…In order to do this a common technique consists of tracking each point across consecutive frames and include only the points that can be detected in p consecutive frames [14,7]. As a result, the number of landmarks in the map is reduced and also the complexity of the SLAM problem.…”

Section: Evaluation Criteriamentioning

confidence: 99%

“…Finally, the quality of the results should be analyzed to evaluate different detectors and descriptors. In some cases, the path of the robot estimated using a visual SLAM approach is compared to the estimation using laser range data [7,29], since laser-based SLAM usually produces more accurate results. However, the visual SLAM results greatly depend on the SLAM algorithm used and several implementation details.…”

Section: Introductionmentioning

confidence: 99%

“…The camera maintains always the same position with respect to the robot and it rotates only over the vertical axis. This assumption has been applied to many SLAM approaches [7,24,29]. We additionally assume that the same landmark, when rotated, converts into a different one.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A comparative evaluation of interest point detectors and local descriptors for visual SLAM

Gil

Mozos

Ballesta

2009

Machine Vision and Applications

Self Cite

150

View full text Add to dashboard Cite

In this paper we compare the behavior of different interest points detectors and descriptors under the conditions needed to be used as landmarks in visionbased simultaneous localization and mapping (SLAM). We evaluate the repeatability of the detectors, as well as the invariance and distinctiveness of the descriptors, under different perceptual conditions using sequences of images representing planar objects as well as 3D scenes. We believe that this information will be useful when selecting an appropriate landmark detector and descriptor for visual SLAM.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Siftmentioning

confidence: 99%

Section: Evaluation Criteriamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A comparative evaluation of interest point detectors and local descriptors for visual SLAM

Gil

Mozos

Ballesta

2009

Machine Vision and Applications

Self Cite

150

View full text Add to dashboard Cite

show abstract

CI‐Graph simultaneous localization and mapping for three‐dimensional reconstruction of large and complex environments using a multicamera system

Piniés

Paz

Gálvez-López

et al. 2010

Journal of Field Robotics

View full text Add to dashboard Cite

Submapping and graphical methods have been shown to be valuable approaches to simultaneous localization and mapping (SLAM), providing significant advantages over the classical extended Kalman filter (EKF) solution: they are faster and, when using local coordinates, produce more consistent estimates. The main contribution of this paper is CI-Graph SLAM, a novel algorithm that is able to efficiently map large environments by building a graph of submaps and a spanning tree of this graph with the following properties: (1) any pair of neighboring submaps in the spanning tree are conditionally independent and (2) the current submap is always up to date, containing the marginal probabilities of the submap variables given all previous measurements. Thanks to these properties, an old submap can be updated at any time by performing a single propagation from the current map to the old submap along the spanning tree. This operation is required only when a map is revisited, with a cost linear with the number of maps in the loop. At the end of the experiment the method performs a single propagation through the whole tree, recovering exactly the same marginals for all the map variables as the EKF-SLAM algorithm does, without ever needing to compute the whole covariance matrix. To evaluate CI-Graph performance in extremely loopy environments, the method was tested using a synthetic Manhattan world. The behavior of the algorithm in large real environments is shown using the public data sets from the RAWSEEDS project in which a robot equipped with a trinocular camera traversed indoor and outdoor environments with several loops and revisited areas. Loops are robustly closed using a novel technique that detects candidate loop closures using a visual vocabulary tree and filters them using temporal and geometric constraints. Our experiments show that when using frontal cameras, the technique outperforms FAB-MAP. The epipolar geometry of the loop-closing images is used to find feature matches that are imposed on the CI-Graph to correct the submap estimations along the loop. C

show abstract