One shot learning gesture recognition from RGBD images

Di Wu; Fan Zhu; Ling Shao

doi:10.1109/CVPRW.2012.6239179

One shot learning gesture recognition from RGBD images

Di Wu, Fan Zhu, Ling Shao

School of Computer Science

Research output: Contribution to conference › Paper

129 Citations (Scopus)

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.

Original language	English
DOIs	https://doi.org/10.1109/CVPRW.2012.6239179
Publication status	Published - Jun 2012
Event	CVPRW 2012 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops - Providence, USA Duration: 1 Jun 2012 → …

Conference

Conference	CVPRW 2012 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
Period	1/06/12 → …

Keywords

feature extraction
gesture recognition
image classification
image colour analysis
image denoising
image segmentation
artificial intelligence
spatial variables measurement

Access to Document

10.1109/CVPRW.2012.6239179

Cite this

@conference{ad84c044a42041fa830fc6e2d943474f,

title = "One shot learning gesture recognition from RGBD images",

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.",

keywords = "feature extraction, gesture recognition, image classification, image colour analysis, image denoising, image segmentation, artificial intelligence, spatial variables measurement",

author = "Di Wu and Fan Zhu and Ling Shao",

year = "2012",

month = jun,

doi = "10.1109/CVPRW.2012.6239179",

language = "English",

note = "CVPRW 2012 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops ; Conference date: 01-06-2012",

}

TY - CONF

T1 - One shot learning gesture recognition from RGBD images

AU - Wu, Di

AU - Zhu, Fan

AU - Shao, Ling

PY - 2012/6

Y1 - 2012/6

N2 - 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.

AB - 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.

KW - feature extraction

KW - gesture recognition

KW - image classification

KW - image colour analysis

KW - image denoising

KW - image segmentation

KW - artificial intelligence

KW - spatial variables measurement

UR - https://www.scopus.com/pages/publications/84865029506

U2 - 10.1109/CVPRW.2012.6239179

DO - 10.1109/CVPRW.2012.6239179

M3 - Paper

T2 - CVPRW 2012 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops

Y2 - 1 June 2012

ER -

One shot learning gesture recognition from RGBD images

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this