سال انتشار: ۱۳۸۵
محل انتشار: دوازدهمین کنفرانس سالانه انجمن کامپیوتر ایران
تعداد صفحات: ۸
Hossein Ebrahimnezhad – Department of Electrical and Computer Engineering Tarbiat Modares University, Tehran, Iran
Hassan Ghassemian – Department of Electrical and Computer Engineering Tarbiat Modares University, Tehran, Iran
Profiles of an object provide rich information about its geometry and can be used for shape reconstruction under known object movement. In this paper we propose a method to estimate the motion from space curves. The moving object is captured by two cameras during time. A robust curve stereo matching algorithm is employed to extract the precise location of convinced space curves for any sequence. By tracking of the space curves, the rigid motion estimation of object can be accomplished. So, by unifying the large number of virtual cameras, the rays constructed from silhouette are intersected to recover the fine visual hull. Because of using the curve matching scheme instead of color matching, the proposed method is less sensitive to color adjustment between cameras and illumination changes of light source. Our method is applicable also to the low-texture object.