摘要
提出一种从图像序列重建三维物体多面体的算法,与传统算法相比,具有数据冗余度较低、精度较高、健壮性较好等优点.实验结果表明,使用该算法计算、生成的物体可视外壳更加逼近真实物体.
In this paper,a new algorithm for reconstructing a visual hull of an object from its silhouette is proposed. Compared with the traditional methods,this algorithm has lower degree of data redundancy and higher precision,robustness,etc. The experiments show that the result of reconstruction is better approximation of the real objects.
出处
《河南大学学报（自然科学版）》
CAS
北大核心
2010年第3期303-306,共4页
Journal of Henan University:Natural Science
基金
河南省重点攻关项目(08210224003)
关键词
多视角
可视外壳
三维重建
线段相交
multi-angles of view
visual hull
3-D reconstruction
segment intersection
