Object Tracking Using Adaptive Object Color Modeling

This paper presents an object tracking method based on the object and background color. A three dimensional feature vector is used to describe color feature (R, G, B) of each pixel. A quantized histogram of that feature value of pixels on the selected object and a quantized histogram of pixels from the neighborhood surrounding the object are developed. The RGB based joint probability density function of the selected object region and that of a neighborhood surrounding the object are used to separate object from background. An adaptive color model of separated object is developed to detect the object in subsequent frames while Mean-shift procedure is used to track the object. The proposed tracker is illustrated to be able to track the object and successfully resolve the problems caused by the camera movement and illumination change, rotation, shape deformation and 3D transformation of the target object.