One shot learning gesture recognition from RGBD images

Author

Wu, Di ; Zhu, Fan ; Shao, Ling

Author_Institution

Univ. of Sheffield, Sheffield, UK

fYear

2012

fDate

16-21 June 2012

Firstpage

7

Lastpage

12

Abstract

We present a system to classify the gesture from only one learning example. The inputs are duo-modality, i.e. RGB and depth sensor from Kinect. Our system performs morphological denoising on depth images and automatically segments the temporal boundaries. Features are extracted based on Extended-Motion-History-Image (Extended-MHI) and the Multi-view Spectral Embedding (MSE) algorithm is used to fuse duo modalities in a physically meaningful manner. Our approach achieves less than 0.3 in Levenshtein distance in CHALEARN Gesture Challenge validation batches [1].

Keywords

feature extraction; gesture recognition; image classification; image colour analysis; image denoising; image segmentation; learning (artificial intelligence); spatial variables measurement; CHALEARN gesture challenge validation batch; Kinect RGB sensor; Kinect depth sensor; Levenshtein distance; MSE algorithm; RGBD images; automatic temporal boundary segmentation; depth image denoising; duo-modality; extended-MHI algorithm; extended-motion-history-image algorithm; feature extraction; gesture classification; morphological denoising; multiview spectral embedding algorithm; one shot learning gesture recognition; Cameras; Image segmentation; Motion segmentation; Noise; Noise reduction; Training; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Vision and Pattern Recognition Workshops (CVPRW), 2012 IEEE Computer Society Conference on

Conference_Location

Providence, RI

ISSN

2160-7508

Print_ISBN

978-1-4673-1611-8

Electronic_ISBN

2160-7508

Type

conf

DOI

10.1109/CVPRW.2012.6239179

Filename

6239179