Improving Visual Object Tracking Using General UFIR and Kalman Filters Under Disturbances in Bounding Boxes

A well-known problem of visual object tracking is the difficulty of accurately estimating the object trajectory under conditions of environmental disturbances in the bounding box (BB) of a video camera. In this paper, we consider BB variations as Gaussian-Markov colored measurement noise (CMN). In order to perform accurate tracking in the presence of CMN, we use measurement differencing and develop a robust general unbiased finite impulse response (GUFIR) filter and use the general Kalman filter (GKF) as a benchmark. The GUFIR and GKF algorithms are tested by the Car4benchmark. It is shown that, in terms of the tracking precision and under the heavy disturbance with the 0.65 = ? = 0.95 coloredness factor, the best tracking performance is achieved using the robust GUFIR filter. When ? < 0.6, the GUFIR and GKF algorithms perform near equally. In the extreme point of ? = 1.0, where the Gauss-Markov CMN loses the stationarity, both algorithms provide zero precision and become inefficient. In general, it is concluded that the GUFIR filter, which ignores any zero mean disturbance and initial values, is much more suitable for applications in visual object tracking than Kalman-like algorithms relying on complete object information.

Automated summary

The paper addresses the challenge of accurately estimating object trajectory in the bounding box of a video camera under environmental disturbances in visual object tracking. It considers bounding box variations as Gaussian-Markov colored measurement noise and proposes a robust unbiased finite impulse response filter as well as a general Kalman filter as a benchmark. The Car4benchmark is used to test the algorithms, and the results show that the robust GUFIR filter achieves the best tracking performance under heavy disturbances.