Abstract
Re-identification is commonly accomplished using appearance features based on salient points and color information. In this paper, we make an study on the use of different features exclusively obtained from depth images captured with RGB-D cameras. The results achieved, using simple geometric features extracted in a top-view setup, seem to provide useful descriptors for the re-identification task. © 2012 Springer-Verlag.
Author supplied keywords
Cite
CITATION STYLE
Lorenzo-Navarro, J., Castrillón-Santana, M., & Hernández-Sosa, D. (2012). An study on re-identification in RGB-D imagery. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7657 LNCS, pp. 200–207). https://doi.org/10.1007/978-3-642-35395-6_28
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
