计算机应用研究2025，Vol.42Issue(1)：307-313,7.DOI:10.19734/j.issn.1001-3695.2024.01.0111
一种基于旋转-平移解耦优化的在线稠密重建算法
Novel method for online dense reconstruction by rotation-translation decoupling optimization
摘要
Abstract
To address the optimization instability caused by the rotation-translation-coupling-optimization in traditional algo-rithms,this paper proposed a novel dense reconstruction algorithm based on rotation-translation decoupling optimization.The algorithm divided the camera pose space into independent rotation and translation subspaces and independently searches for the optimal solution within each subspace.During each iteration,the algorithm set search neighborhoods for the estimated values of rotation and translation,samples candidate solutions within these neighborhoods,and selected the optimal solutions through evaluation to update the estimates.Iterate this process until the termination condition was met.Experimental results verify that the algorithm effectively reduces interference between rotation and translation during the optimization process,thereby enhan-cing the efficiency and accuracy of the optimization.It demonstrates that the algorithm has certain advantages in the camera pose estimation stage of dense reconstruction.关键词
在线稠密重建/解耦优化/相机位姿估计/隐式曲面配准Key words
online dense reconstruction/decoupling optimization/camera pose estimation/implicit surface registration分类
信息技术与安全科学引用本文复制引用
郭帆,吕泽均,张严辞..一种基于旋转-平移解耦优化的在线稠密重建算法[J].计算机应用研究,2025,42(1):307-313,7.基金项目
国家重大专项资助项目(GJXM92579) （GJXM92579）
四川省重点研发项目(2023YFG0122) （2023YFG0122）
