Body-relative navigation guidance using uncalibrated cameras

Koch, Olivier; Teller, Seth

doi:10.1109/iccv.2009.5459327

Cited by 8 publications

(5 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A final relevant paper is the recent work of Koch and Teller [12], who automatically learn about how feature motion in the images of a multi-camera rig relates to the rig's motion as it is carried by a person, but stop short of aiming to determine full geometric extrinsic calibration.…”

Section: Related Workmentioning

confidence: 99%

SLAM-based automatic extrinsic calibration of a multi-camera rig

Carrera

Angeli

Davison

2011

2011 IEEE International Conference on Robotics and Automation

111

View full text Add to dashboard Cite

Abstract-Cameras are often a good choice as the primary outward-looking sensor for mobile robots, and a wide field of view is usually desirable for responsive and accurate navigation, SLAM and relocalisation. While this can potentially be provided by a single omnidirectional camera, it can also be flexibly achieved by multiple cameras with standard optics mounted around the robot. However, such setups are difficult to calibrate.Here we present a general method for fully automatic extrinsic auto-calibration of a fixed multi camera rig, with no requirement for calibration patterns or other infrastructure, which works even in the case where the cameras have completely non-overlapping views. The robot is placed in a natural environment and makes a set of programmed movements including a full horizontal rotation and captures a synchronized image sequence from each camera. These sequences are processed individually with a monocular visual SLAM algorithm. The resulting maps are matched and fused robustly based on corresponding invariant features, and then all estimates are optimised full joint bundle adjustment, where we constrain the relative poses of the cameras to be fixed. We present results showing accurate performance of the method for various two and four camera configurations.

show abstract

Section: Related Workmentioning

confidence: 99%

SLAM-based automatic extrinsic calibration of a multi-camera rig

Carrera

Angeli

Davison

2011

2011 IEEE International Conference on Robotics and Automation

111

View full text Add to dashboard Cite

show abstract

“…Discussion: In general, most methods to detect human motion have been designed to employ passive and wearable approaches. These methods have proven to be effective in representing the action that takes place in the scene [13], [14]; unfortunately, they cannot interpret user intention (defined as the reach-to-grasp action towards an object) because they are not designed to capture the user's visual system. For example, the system proposed by Sakita, et al [5] exploits human gaze to support cooperative work with robots.…”

Section: Related Workmentioning

confidence: 98%

Prediction of user's grasping intentions based on eye-hand coordination

Carrasco

Clady

2010

2010 IEEE/RSJ International Conference on Intelligent Robots and Systems

View full text Add to dashboard Cite

Eye-hand coordination is a primordial reach-tograsp action performed by a human hand when reaching for an object. This paper proposes the use of a visual sensor which allows the simultaneous analysis of hand and eye motions in order to recognize the reach-to-grasp movement, i.e. to predict the grasping gesture. This solution fuses two viewpoints taken from the user's perspective. First, by using an eye-tracker device attached to the user's head; and second, by utilizing a wearable camera attached to the user's hand. The information from these two viewpoints is used to characterize multiple hand movements in conjunction with eye-gaze movements through a HiddenMarkov Model framework. In various experiments, we show that combining these two sources of information allows the prediction of a reach-to-grasp movement as well as the desired object.

show abstract

“…Another related approach is the vision based navigation assistant presented in [11], which builds a topological map using a wide field of view system composed of four cameras mounted on the user's shoulder.…”

Section: Yes Nomentioning

confidence: 99%

Wearable omnidirectional vision system for personal localization and guidance

Murillo

Gutiérrez-Gómez

Rituerto

et al. 2012

2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops

View full text Add to dashboard Cite

Autonomous navigation and recognition of the environment are fundamental abilities for people extensively studied in computer vision and robotics fields. Expansion of low cost wearable sensing provides interesting opportunities for assistance systems that augment people navigation and recognition capabilities. This work presents our wearable omnidirectional vision system and a novel twophase localization approach running on it. It runs stateof-the-art real time visual odometry adapted to catadioptric images augmented with topological-semantic information. The presented approach benefits from using wearable sensors to improve visual odometry results with true scaled solution. The wide field of view of catadioptric vision system used makes features last longer in the field of view and allows more compact location representation which facilitates topological place recognition. Experiments in this paper show promising ego-localization results in realistic settings, providing good true scaled visual odometry estimation and recognition of indoor regions.

show abstract

Body-relative navigation guidance using uncalibrated cameras

Abstract: We present a vision-based

Cited by 8 publications

References 22 publications

SLAM-based automatic extrinsic calibration of a multi-camera rig

SLAM-based automatic extrinsic calibration of a multi-camera rig

Prediction of user's grasping intentions based on eye-hand coordination

Wearable omnidirectional vision system for personal localization and guidance

Contact Info

Product

Resources

About