Automatic Seamless Video Mosaic from Webcams Using LSF Techniques

Abstract

This paper presents an automatic technique to create a wide field-of-view from common webcams. We eliminate the manual selection of corresponding pixels where no prior information is required. Our system consists of three stages: In the first stage, we detect robust features in the initial frame of each webcam and find the corresponding points between them. Then the matched point pairs are employed to compute the perspective matrix which describes the geometric relationship of the adjacent views. After the initialization stage, we register the frame sequences of different webcams on the same plane using the perspective matrix and synthesize the overlapped region using a nonlinear blending method in real time. In the third stage, due to the existence of seam artifacts, exposure differences and other misalignments along overlap region, we apply least square fitting technique to achieve a wide seamless video mosaic. We demonstrate the effectiveness of our method on a prototype that consists of two ordinary webcams and shows that this is an interesting and inexpensive way to experience the wide-angle observation.