【作者】 陈宁； 郑江滨；

【Author】 Chen Ning, Zheng Jiangbin School of Computer Science, Northwestern Polytechnical University, Xi’an 710072, China

【机构】 西北工业大学计算机学院；

【摘要】 本文利用基于图像绘制的体元重建技术获取场景中目标的三维外形。首先将场景空间离散为体元,然后在CIELab颜色空间判断体元在输入图像上反向投影像素集的颜色一致性,以获取目标的三维信息;最后采用表面修复算法对重建模型表面空洞进行填补。实验结果表明该方法可重建目标外形,并且颜色纹理信息准确,真实感强。更多还原

【Abstract】 In this paper, 3D shape of the object in a real scene is rendered using the voxel reconstruction algorithm based on images.Firstly, initialize the scene as a volume which consists of voxels.Secondly,to extract 3D information of the objects,we have to check the color consistency of the projection pixels in CIELab color space. Finally,a surface recovering algorithm is proposed to fill the holes on the surface of the reconstructed model.Some experimental results demonstrate the algorithm is more efficient to reconstruct the realistic models with detailed appearance.更多还原