摘要
针对基于图象的三维建模方法存在着不精确的问题,提出了一种基于误差最小化的视频自动三维建模方法.该方法利用互信息计算三维模型的拟合得分,并对模型拟合进行优化.该优化方案在不需要人工干预的情况下最小化拟合误差,以提高三维动画图象建模的精度.实验评估部分使用了真实的视频作为数据集,实验结果验证了本文方法的有效性.
Aiming at the problem of inaccuracy of image-based three-dimensional modeling method, a video automatic three-dimensional modeling method based on error minimization is proposed. This method uses mutual information to calculate the fitting score of three-dimensional model, and optimizes the model fitting. The optimization scheme minimizes the fitting error without manual intervention to improve the accuracy of three-dimensional animation image modeling. The experimental evaluation part uses real video as data set. The experimental results verify the effectiveness of the proposed method in this paper.
引文
[1] CHUNG D H S,PARRY M L,LEGG P A,et al.Visualizing multiple error-sensitivity fields for single camera positioning[J].Computing & Visualization in Science,2012,15(6):303-317.
[2] SZELISKI R.Computer vision:algorithms and applications[M].Springer-Verlag New York,Inc,2010.
[3] PANDEY G,MCBRIDE J R,SAVARESE S,et al.Automatic targetless extrinsic calibration of a 3D lidar and camera by maximizing mutual information[C]//Twenty-Sixth AAAI Conference on Artificial Intelligence.AAAI Press,2012:2053-2059.
[4] PANINSKI L.Estimation of entropy and mutual information[J].Neural Computation,2014,15(6):1191-1253.
[5] DAN G,PURNOMO B,SHREINER D,et al.OpenGL ES 3.0programming guide[M].Addison-Wesley Professional,2014.