CPL - Chalmers Publication Library
| Utbildning | Forskning | Styrkeområden | Om Chalmers | In English In English Ej inloggad.

Head Pose Classification by Multi-Class AdaBoost with Fusion of RGB and Depth Images

Yixiao Yun (Institutionen för signaler och system, Signalbehandling) ; Mohamed Hashim Changrampadi (Institutionen för signaler och system, Signalbehandling) ; Irene Y.H. Gu (Institutionen för signaler och system, Signalbehandling)
1st International Conference on Signal Processing and Integrated Networks (SPIN) p. 174-177. (2014)
[Konferensbidrag, refereegranskat]

This paper addresses issues in multi-class visual object classification, where sequential learning and sensor fusion are exploited in a unified framework. We adopt a novel method for head pose classification using RGB and depth images. The main contribution of this paper is a multi-class AdaBoost classification framework where information obtained from RGB and depth modalities interactively complement each other. This is achieved by learning weak hypotheses for RGB and depth modalities independently with the same sampling weight in the boosting structure, and then fusing them through learning a sub-ensemble. Experiments are conducted on a Kinect RGB-D face image dataset containing 4098 face images in 5 different poses. Results have shown good performance in obtaining high classification rate (99.76%) with low false alarms on the dataset.

Nyckelord: head pose classification, visual information fusion, RGB and depth images, AdaBoost



Den här publikationen ingår i följande styrkeområden:

Läs mer om Chalmers styrkeområden  

Denna post skapades 2014-05-14. Senast ändrad 2015-07-06.
CPL Pubid: 198043