Panoramic Background Generation using Mean - Shift in Moving Camera Environment

In this paper, we present a novel method to generate a panoramic background image from a sequence of moving camera images. Our method is based on nonparametric statistics and robust model estimation using meanshift. Each frame is projected to panoramic space by homography matrix which is calculated by SURF(Speeded Up Robust Features) matching. To generate reliable panoramic background image, we use mean-shift. Our proposed method allows the generation of very clear panoramic backgrounds in moving camera environment. Experimental results from real sequences demonstrate the effectiveness of our method Keyword : background generation, mean-shift, background modeling, object detection, obect tracking