Abstract | ||
---|---|---|
In this paper we propose a novel spatially stratified sampling technique for evaluating the likelihood function in particle filters. In particular, we show that in the case where the measurement function uses spatial correspondence, we can greatly reduce computational cost by exploiting spatial structure to avoid redundant computations. We present results which quantitatively show that the technique permits equivalent, and in some cases, greater accuracy, as a reference point cloud particle filter at significantly faster run-times. We also compare to a GPU implementation, and show that we can exceed their performance on the CPU. In addition, we present results on a multi-target tracking application, demonstrating that the increases in efficiency permit online 6DoF multi-target tracking on standard hardware. |
Year | DOI | Venue |
---|---|---|
2015 | 10.1109/WACV.2015.24 | Applications of Computer Vision |
Keywords | Field | DocType |
particle filtering (numerical methods),signal sampling,target tracking,gpu implementation,novel spatially stratified correspondence sampling,online 6dof multitarget tracking,real-time point cloud tracking,reference point cloud particle filter,spatial correspondence,vectors,accuracy,visualization,computational modeling,solid modeling | Computer vision,Central processing unit,Likelihood function,Time point,Computer science,Particle filter,Solid modeling,Artificial intelligence,Sampling (statistics),Computation,Cloud computing | Conference |
ISSN | Citations | PageRank |
2472-6737 | 0 | 0.34 |
References | Authors | |
13 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jeremie Papon | 1 | 199 | 10.18 |
Markus Schoeler | 2 | 145 | 5.98 |
Florentin Wörgötter | 3 | 1304 | 119.30 |