Proceedings of 1995 IEEE International Conference on Robotics and Automation
DOI: 10.1109/robot.1995.525522
|View full text |Cite
|
Sign up to set email alerts
|

Automatic camera placement for robot vision tasks

Abstract: Remote sensors such as CCD cameras can be used for a variety of robot sensing tasks, but given restrictions on camera location and imaging geometry, task constraints, and visual occlusion it can be difficult to find viewing positions from which the task can be completed. The complexity of these constraints suggests that automated, quantitative methods of sensor placement are likely to be useful, particularly when the workspace is cluttered and a mobile robot-mounted sensor is being used to increase the sensibl… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
27
0

Publication Types

Select...
5
3

Relationship

0
8

Authors

Journals

citations
Cited by 46 publications
(27 citation statements)
references
References 11 publications
0
27
0
Order By: Relevance
“…Traditional methods to deal with occlusions as suggested in [14] are to use multiple cameras or to predict the movement of the target using a track memory containing the history of the previous locations of the target. Another approach reported in [20] includes automatic camera placement that increases the feasible region, circumvents occlusions, and provides uninterrupted tracking. Alternative to pure vision or image processing techniques is a variety of optical flow algorithms [7,[21][22][23] addressing the task of 3-D motion reconstruction from the fast sampled 2-D image samples.…”
Section: Doi: 102514/133206mentioning
confidence: 99%
“…Traditional methods to deal with occlusions as suggested in [14] are to use multiple cameras or to predict the movement of the target using a track memory containing the history of the previous locations of the target. Another approach reported in [20] includes automatic camera placement that increases the feasible region, circumvents occlusions, and provides uninterrupted tracking. Alternative to pure vision or image processing techniques is a variety of optical flow algorithms [7,[21][22][23] addressing the task of 3-D motion reconstruction from the fast sampled 2-D image samples.…”
Section: Doi: 102514/133206mentioning
confidence: 99%
“…Triggs et al [38] gave a method of function optimization technique to minimize their viewpoint evaluation function. They divided the search space into a set of local regions and built a probabilistic function interpolation.…”
Section: A Previous Approachesmentioning
confidence: 99%
“…It raises the problem of autonomous exploration [6], [8], [16], [20], [23], [22]. In [8], the sensor (a range finder) placement is computed from a local map of the scene that is described by an octree.…”
Section: Related Workmentioning
confidence: 99%
“…Indeed this is not realistic, in practice, owing to the size of the configuration space. As in [19], [20] we have thus defined a function to be minimized that integrates the constraints imposed by the robotic system and evaluates the quality of the viewpoint. The function ) to be optimized is taken as a weighted sum of a set of measures that determine the quality or the badness of a viewpoint.…”
Section: Viewpoint Selectionmentioning
confidence: 99%
See 1 more Smart Citation