Simulation videos for understanding occlusion effects on kernel based object tracking

• Published in: Lecture Notes in Electrical Engineering
• Main Author: Lee B.Y.; Liew L.H.; Cheah W.S.; Wang Y.C.
• Format: Conference paper
• Language: English
• Published: 2012
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-84868586856&doi=10.1007%2f978-94-007-5699-1_15&partnerID=40&md5=6b2f193e406ad9c91cf3a9d9937798cd

Occlusion handling is one of the most studied problems for object tracking in computer vision. Many previous works claimed that occlusion can be handled effectively using Kalman filter, Particle filter and Mean Shift tracking methods. However, these methods were only tested on specific task videos. In order to explore the actual potential of these methods, this paper introduced 64 simulation video sequences to experiment the effectiveness of each tracking methods on various occlusion scenarios. Tracking performances are evaluated based on Sequence Frame Detection Accuracy (SFDA). The results showed that Mean shift tracker would fail completely when full occlusion occurred. Kalman filter tracker achieved highest SFDA score of 0.85 when tracking object with uniform trajectory and no occlusion. Results also demonstrated that Particle filter tracker fails to detect object with non-uniform trajectory. The effect of occlusion on each tracker is analyzed with Frame Detection Accuracy (FDA) graph. © 2012 Springer Science+Business Media.