摘要
提出一种基于光流反馈的单目视觉三维(3D)重建方法,实现对场景快速、准确的3D立体化建模。由帧间光流场建立更为稳健的同名像点匹配关系,同时运用五点算法估计摄像机的相对位姿,以构建稀疏点云和初始网格。从运动视觉分析的角度寻求多视重构的求解方法,将重建模型反馈至重建过程,用各视图像的偏差驱动模型变形。将粗略、不准确的原始网格曲面经过致密的非刚性变形,调整至精确的曲面。在统一计算设备架构下,利用图形处理器对光流算法进行并行加速,显著提高了重构算法运行的实时性。室内真实场景下的重建结果证明了所提算法的可行性与准确性。
A monocular three dimensional（3D） reconstruction technique based on optical flow feedback is proposed to achieve fast and accurate 3D stereoscopic modeling in the real scene. Corresponding pixel pairs are robustly matched by inter-frame optical flow fields and the five-point algorithm is employed to determine relative pose of the moving camera, therefore sparse point cloud is generated and initial crude mesh is built. In the proposed method,multi-view reconstruction is implemented from perspective of vision method on motion analysis. The reconstruction model is fed-back to the reconstruction process and the model is deformed by utilizing the bias-driven of each view.The coarse and inaccurate original mesh surface is adjusted to the exact surface through a dense non-rigid deformation. Under the compute unified device architecture, the optical flow algorithm is optimized in parallel mode by using the graphic processing unit hardware and real-time performance of the reconstruction algorithm is significantly improved. The experimental results obtained in realistic indoor scenario demonstrate the effectiveness and accuracy of the proposed algorithm.
出处
《光学学报》
EI
CAS
CSCD
北大核心
2015年第5期228-236,共9页
Acta Optica Sinica
基金
国家自然科学基金(61105033
61175087)
关键词
机器视觉
三维重建
光流
场景流
统一计算设备架构
machine vision three dimensional reconstruction optical flow scene flow compute unified device architecture
作者简介
李秀智（1979-），男，博士，讲师，主要从事智能机器人导航、机器视觉等方面的研究。E—mail：xiuzhi．1ee@163．com。
通信联系人。E-mail：yanglinligong@sina．com
