Panoramic Vision and Laser Range Finder Fusion for Multiple Person Tracking

Chakravarty, Punarjay; Jarvis, Ray

doi:10.1109/iros.2006.282149

Cited by 41 publications

(35 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The standard Kalman filter [3] provides an efficient way to integrate different sensor data and, in case of linear systems with Gaussian noise, it is known to be optimal. An Extended Kalman Filter (EKF) can be used to provide approximate solutions in case of non-linearities, although most of the recent approaches for tracking people are based on particle filters [14,39] because they are not constrained by any linear or Gaussian assumption. Unfortunately, in terms of computational cost, particle filters can be very demanding and pose serious constraints in case of BoFs or multiple people tracking.…”

Section: Methodsmentioning

confidence: 99%

“…In [39], the authors illustrate a robot equipped with two laser range sensors that can track several people using a combination of particle filters and probabilistic data association. Another solution based on particle filter is proposed in [14], which integrates laser data and visual information from a panoramic camera. A covariance intersection method, using sonar, laser and visual data, is implemented in [31] for tracking multiple people.…”

Section: Related Workmentioning

confidence: 99%

“…Most of the existing legs detection algorithms are based on the search of local minima [9,39], motion detection [14,29] or machine learning techniques [1].…”

Section: Legs Detectionmentioning

confidence: 99%

“…The face recognition measure from Eigenface is modeled as follows: (14) where ξ(c i ) is the value obtained for the ith face in the database in case of perfect match (ξ(c i ) = −1). For simplicity, the noise n ξ k is assumed to be normally distributed, with standard deviation determined by a number of tests with different faces.…”

Section: Observation Models Of the Recognizersmentioning

confidence: 99%

See 3 more Smart Citations

A Bank of Unscented Kalman Filters for Multimodal Human Perception with Mobile Service Robots

Bellotto

2010

Int J of Soc Robotics

View full text Add to dashboard Cite

A new generation of mobile service robots could be ready soon to operate in human environments if they can robustly estimate position and identity of surrounding people. Researchers in this field face a number of challenging problems, among which sensor uncertainties and real-time constraints. In this paper, we propose a novel and efficient solution for simultaneous tracking and recognition of people within the observation range of a mobile robot. Multisensor techniques for legs and face detection are fused in a robust probabilistic framework to height, clothes and face recognition algorithms. The system is based on an efficient bank of Unscented Kalman Filters that keeps a multi-hypothesis estimate of the person being tracked, including the case where the latter is unknown to the robot. Several experiments with real mobile robots are presented to validate the proposed approach. They show that our solutions can improve the robot's perception and recognition of humans, providing a useful contribution for the future application of service robotics.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

“…Most of the existing legs detection algorithms are based on the search of local minima [9,39], motion detection [14,29] or machine learning techniques [1].…”

Section: Legs Detectionmentioning

confidence: 99%

Section: Observation Models Of the Recognizersmentioning

confidence: 99%

See 2 more Smart Citations

A Bank of Unscented Kalman Filters for Multimodal Human Perception with Mobile Service Robots

Bellotto

2010

Int J of Soc Robotics

View full text Add to dashboard Cite

show abstract

“…Both the devices have been used in (Bellotto and Hu 2009) applying sensor fusion techniques and UKF estimation to perform people tracking in typical office environments. Several other approaches have been proposed using particle filters with laser data and/or vision (Schulz et al 2003a;Chakravarty and Jarvis 2006).…”

mentioning

confidence: 99%

Computationally efficient solutions for tracking people with a mobile robot: an experimental evaluation of Bayesian filters

Bellotto

2009

Auton Robot

View full text Add to dashboard Cite

Modern service robots will soon become an essential part of modern society. As they have to move and act in human environments, it is essential for them to be provided with a fast and reliable tracking system that localizes people in the neighbourhood. It is therefore important to select the most appropriate filter to estimate the position of these persons. This paper presents three efficient implementations of multisensor-human tracking based on different Bayesian estimators: Extended Kalman Filter (EKF), Unscented Kalman Filter (UKF) and Sampling Importance Resampling (SIR) particle filter. The system implemented on a mobile robot is explained, introducing the methods used to detect and estimate the position of multiple people. Then, the solutions based on the three filters are discussed in detail. Several real experiments are conducted to evaluate their performance, which is compared in terms of accuracy, robustness and execution time of the estimation. The results show that a solution based on the UKF can perform as good as particle filters and can be often a better choice when computational efficiency is a key issue.

show abstract