Geometric particle swarm optimization for robust visual ego-motion estimation via particle filtering

作者：

Highlights：

• Novel particle filtering based visual ego-motion estimation algorithm robust to abrupt camera motion.

• Multi-layered importance sampling via particle swarm optimization (PSO)

• Reformulation of the conventional vector space PSO algorithm by geometric special Euclidean group SE(3)

• Efficient convergence of PSO and real-time visual ego-motion estimation performance

摘要

•Novel particle filtering based visual ego-motion estimation algorithm robust to abrupt camera motion.•Multi-layered importance sampling via particle swarm optimization (PSO)•Reformulation of the conventional vector space PSO algorithm by geometric special Euclidean group SE(3)•Efficient convergence of PSO and real-time visual ego-motion estimation performance

论文关键词：Visual ego-motion estimation,Visual odometry,Visual SLAM,Particle filtering,Particle swarm optimization,Special Euclidean group

论文评审过程：Received 13 June 2012, Revised 20 March 2013, Accepted 25 April 2013, Available online 4 May 2013.

论文官网地址：https://doi.org/10.1016/j.imavis.2013.04.004