BROWSE
Detail
Conference Paper
Person Re-identification in a Person-following Robot
Cited - time in 
Share 
Share
- Authors
- Yongwoo Yoon, Hosub Yoon, Jaehong Kim
- Issue Date
- 2016-08
- Citation
- International Symposium on Robot and Human Interactive Communication (RO-MAN) 2016, pp.387-388
- Language
- English
- Type
- Conference Paper
- Abstract
- Person re-identification can often be a problem in person-following robots under the environment of severe occlusions. This paper improves the person re-identification through two different approaches. The first one is an integration of global and local metrics for feature matching. The local metric is learned from multiple shots of a person. Secondly, physical search spaces for the person re-identification are restricted by estimating the position of a target person. A preliminary experiment shows improvements over existing methods.
- KSP Keywords
- Feature matching, Global and local, Local metrics, Person Re-Identification, Person-following robot