3D RECONSTRUCTION USING PARTICLE SWARM OPTIMIZATION
In this paper, a new 3D reconstruction algorithm based on particle swarm optimization (PSO) is proposed. This is the first time to introduce PSO to the multi-views problem. The proposed algorithm designs a scheme to represent the 3D parameters using a particle. Then the PSO algorithm is used to search the solution space and optimizes the fitness function. Finally, the algorithm can efficiently get the correct solution with less time and less complexity. Additionally, some details of implement are discussed which influence the performance of the algorithm. Experiments and comparisons are given in real data. The accuracy of the recovered solution is compared to the existing algorithms and outperformed them.
3D reconstruction Pose estimation Particle swarm optimization Virtual reality
WENHUILI DONGFENG HAN
College of Computer Science and Technology, Key Laboratory of Symbol Computation and Knowledge Engineering of the Ministry of Education, Jilin University, Changchun, 130012, P. R. China
国际会议
开封
英文
406-413
2006-10-15（万方平台首次上网日期，不代表论文的发表时间）