Thumbnail
Access Restriction
Open

Author Papo, A. ♦ Averbuch, A. ♦ Bobrovsky, B. Z.
Source CiteSeerX
Content type Text
File Format PDF
Subject Domain (in DDC) Computer science, information & general works ♦ Data processing & computer science
Subject Keyword Measurement Association Model ♦ Particle Filter ♦ Automatic Tracking Occluded Soccer Player ♦ Joint Probabilistic Data Association Method ♦ Non-linear State-space Model ♦ Contour Measurement ♦ Sequential Monte-carlo ♦ Automatic Analysis ♦ Multi-target Tracking Method ♦ Past Year ♦ Occlusion Occur ♦ Monte Carlo Method ♦ Several Solution ♦ Complete Solution ♦ Different Measurement ♦ Consumer Level ♦ Soccer Game ♦ Sport Player ♦ Measurement Association Method ♦ Player State ♦ Complex Situation ♦ Interactive Digital Tv Spread ♦ Real Soccer Video Sequence ♦ Soccer Video Sequence ♦ Particle Filter Method ♦ Key Word ♦ Multiple Target Tracking
Abstract Tracking sport players in soccer video sequences is very important for automatic analysis of soccer games especially when interactive digital TV spreads to the consumer level. This problem has gained interest among researchers in the past few years. Several solutions have been suggested but still no complete solution was offered, which can automatically track players in scenarios where occlusion occur. A different tracking solution is proposed, which takes the contours that are originated from the players, as an input and produces five different measurements from each contour for estimating the players ’ states. We introduce a measurement association method for associating the contour’s measurements to the players. This method is based on a coupled sample based joint probabilistic data association method (CSBJPDA) , a method which combines between particle filter method, which is a Monte Carlo method that estimates non-Gaussian, non-linear state-space models and joint probabilistic data association method (JPDA), which is a multi-target tracking method. The above method is tested on real soccer video sequences and the results show that it successfully increases the probability to automatically track and distinguish between players even when it is applied on complex situations, where two or more players from the same team occlude one another and abruptly change their directions. 1 key words: occlusion, multiple target tracking, measurement association model, sequential Monte-Carlo 1
Educational Role Student ♦ Teacher
Age Range above 22 year
Educational Use Research
Education Level UG and PG ♦ Career/Technical Study
Publisher Date 2005-01-01