Skip to Main content Skip to Navigation
Conference papers

Combination of RGB-D Features for Head and Upper Body Orientation Classification

Laurent Fitte-Duval 1 Alhayat Ali Mekonnen 1 Frédéric Lerasle 1
1 LAAS-RAP - Équipe Robotique, Action et Perception
LAAS - Laboratoire d'analyse et d'architecture des systèmes
Abstract : In Human-Robot Interaction (HRI), the intention of a person to interact with another agent (robot or human) can be inferred from his/her head and upper body orientation. Furthermore, additional information on the person's overall intention and motion direction can be determined with the knowledge of both orientations. This work presents an exhaustive evaluation of various combinations of RGB and depth image features with different classifiers. These evaluations intend to highlight the best feature representation for the body part orientation to classify, i.e, the person's head or upper body. Our experiments demonstrate that high classification performances can be achieved by combining only three families of RGB and depth features and using a multiclass SVM classifier.
Document type :
Conference papers
Complete list of metadata

Cited literature [24 references]  Display  Hide  Download
Contributor : Frédéric Lerasle <>
Submitted on : Tuesday, April 10, 2018 - 5:34:53 PM
Last modification on : Thursday, June 10, 2021 - 3:04:19 AM


Files produced by the author(s)


  • HAL Id : hal-01763125, version 1


Laurent Fitte-Duval, Alhayat Ali Mekonnen, Frédéric Lerasle. Combination of RGB-D Features for Head and Upper Body Orientation Classification. Advanced Concepts for Intelligent Vision Systems , Oct 2016, Lecce, Italy. ⟨hal-01763125⟩



Record views


Files downloads