MOTION ESTIMATION AND DETECTION OF COMPLEX OBJECT BY ANALYZING RESAMPLED MOVEMENTS OF PARTS (MA-P4)

Author(s) : Punpiti Piamsa-nga (Kasetsart University, Thailand) Noboru Babaguchi (Osaka University, Japan)

Abstract : A moving object that has many complex moving parts is very hard to detect and its motion is not easy to estimate. In this paper, we present a new technique for complex moving object detection using resampled movements of the object’s parts. The Kalman filter is used to track all resampled movements and the tracked routes are classified into groups that share the fundamental movement. Our simulation results show that the recall of complex moving object detection is approximately 0.8, while the computation drops exponentially.

Menu