Interacting multiple model (IMM) Kalman filters for robust high speed human motion tracking

Abstract

Accurate and robust tracking of humans is of growing interest in the inrage processing and computer vision communiries. The a b i l i h of a vision system f o track the subjects and accurately predict their future locations is critical to man? surveillance and camera control applications. Further, an inference of the ppe of motion as 4 1 as to rapidly detect and switch between morion models is critical since in some applications the switching time between motion models can be extremely small. The Interacting Miilriple Model (IMM) Kalnum filter provides a powerful f r a m e w r k for performing the tracking of both the motion as well as the shape of these subjects. The tracking q s t e m utilizes a simple geometric shape primirive such as an ellipse to define a bounding extent of the subject. The utili0 of the IMM paradigm for rapid model switching and behaviour detecrion is shown for a passenger airbag suppression system in an automobile. The simplicity of rhe tnerhods and the robusrness of the underlying IMMfilterIng make the framework well suited for low-cost embedded real-time motion sequence analysis systems.

On-line Support Vector Regression of the transition model for the Kalman filter

“…A celebrated algorithm based on this idea is the Interacting Multiple Model (IMM) filter [18], that has been used also for visual tracking [19,20]. We will show in the experimental results how our approach consistently outperforms IMM, having also the additional advantage to expose the user to less free parameters to be specified.…”

Occupant classification system for automotive airbag suppression

“…Recently, there has been considerable attention paid to developing 'smart' airbags that can determine not only if they should be deployed in a crash event but also with what force they should be deployed [1,2,3,4]. The size and type of occupant needs to be used to determine the safe level of force used to deploy the airbag.…”

“…A wide variety of systems have been proposed for solving this problem [1,2,3,4]. The aim of this paper is to show the applicability of machine vision (using a single B/W camera) to solve the airbag static suppression problem.…”

A Kinect-based system for automatic recording of some pigeon behaviors

“…The covariance information from each model's Kalman filter for beak and head location is used to calculate the model likelihood (Farmer et al, 2002) and combined with a Markov switching matrix, S. The switching matrix controls the selection of which of the two models suits the observed data better. This matrix S ¼ 0:9 0:1 ½ 0:050:95 was chosen to prefer the peck model slightly; this enhances fast peck switching (fast detection of peck), and the greater error for the peck model will quickly cause a transfer back again should the switch not be supported (i.e., should the data not support it).…”

“…An IMM approach was used by Farmer, Hsu, and Jain (2002) for rapid model switching and behavior detection for an airbag suppression system, to distinguish human-initiated versus crash-driven body motions. Although nonpeck and peck motions are both initiated by the pigeon, the difference in velocity of these kinds of motion is similar to the one successfully addressed by Farmer et al This approach has the advantage that the model-switching parameters and filter prediction can be also used to classify the start of a pecking motion and the direction of pecking.…”