Fusion of Range and Intensity Information for View Invariant Gesture Recognition

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

29 Citationer (Scopus)

Resumé

This paper presents a system for view invariant gesture recognition. The approach is based on 3D data from a CSEM SwissRanger SR-2 camera. This camera produces both a depth map as well as an intensity image of a scene. Since the two information types are aligned, we can use the intensity image to define a region of interest for the relevant 3D data. This data fusion improves the quality of the range data and hence results in better recognition. The gesture recognition is based on finding motion primitives in the 3D data. The primitives are represented compactly and view invariant using harmonic shape context. A probabilistic Edit Distance classifier is applied to identify which gesture best describes a string of primitives. The approach is trained on data from one viewpoint and tested on data from a different viewpoint. The recognition rate is 92.9% which is similar to the recognition rate when training and testing on gestures from the same viewpoint, hence the approach is indeed view invariant.
OriginalsprogEngelsk
TitelComputer Vision and Pattern Recognition Workshops, 2008
ForlagElectrical Engineering/Electronics, Computer, Communications and Information Technology Association
Publikationsdato2008
Sider1-7
ISBN (Trykt)9781424423392
StatusUdgivet - 2008
BegivenhedComputer Vision and Pattern Recognition Workshops, 2008. CVPR Workshops 2008. IEEE Computer Society Conference on - Anchorage, Alaska, Canada
Varighed: 23 jun. 200828 jun. 2008

Konference

KonferenceComputer Vision and Pattern Recognition Workshops, 2008. CVPR Workshops 2008. IEEE Computer Society Conference on
LandCanada
ByAnchorage, Alaska
Periode23/06/200828/06/2008

Citer dette

Holte, M. B., Moeslund, T. B., & Fihl, P. (2008). Fusion of Range and Intensity Information for View Invariant Gesture Recognition. I Computer Vision and Pattern Recognition Workshops, 2008 (s. 1-7). Electrical Engineering/Electronics, Computer, Communications and Information Technology Association.
Holte, Michael Boelstoft ; Moeslund, Thomas B. ; Fihl, Preben. / Fusion of Range and Intensity Information for View Invariant Gesture Recognition. Computer Vision and Pattern Recognition Workshops, 2008. Electrical Engineering/Electronics, Computer, Communications and Information Technology Association, 2008. s. 1-7
@inproceedings{dd2b3b506ad111dd92a2000ea68e967b,
title = "Fusion of Range and Intensity Information for View Invariant Gesture Recognition",
abstract = "This paper presents a system for view invariant gesture recognition. The approach is based on 3D data from a CSEM SwissRanger SR-2 camera. This camera produces both a depth map as well as an intensity image of a scene. Since the two information types are aligned, we can use the intensity image to define a region of interest for the relevant 3D data. This data fusion improves the quality of the range data and hence results in better recognition. The gesture recognition is based on finding motion primitives in the 3D data. The primitives are represented compactly and view invariant using harmonic shape context. A probabilistic Edit Distance classifier is applied to identify which gesture best describes a string of primitives. The approach is trained on data from one viewpoint and tested on data from a different viewpoint. The recognition rate is 92.9{\%} which is similar to the recognition rate when training and testing on gestures from the same viewpoint, hence the approach is indeed view invariant.",
author = "Holte, {Michael Boelstoft} and Moeslund, {Thomas B.} and Preben Fihl",
year = "2008",
language = "English",
isbn = "9781424423392",
pages = "1--7",
booktitle = "Computer Vision and Pattern Recognition Workshops, 2008",
publisher = "Electrical Engineering/Electronics, Computer, Communications and Information Technology Association",

}

Holte, MB, Moeslund, TB & Fihl, P 2008, Fusion of Range and Intensity Information for View Invariant Gesture Recognition. i Computer Vision and Pattern Recognition Workshops, 2008. Electrical Engineering/Electronics, Computer, Communications and Information Technology Association, s. 1-7, Computer Vision and Pattern Recognition Workshops, 2008. CVPR Workshops 2008. IEEE Computer Society Conference on, Anchorage, Alaska, Canada, 23/06/2008.

Fusion of Range and Intensity Information for View Invariant Gesture Recognition. / Holte, Michael Boelstoft; Moeslund, Thomas B.; Fihl, Preben.

Computer Vision and Pattern Recognition Workshops, 2008. Electrical Engineering/Electronics, Computer, Communications and Information Technology Association, 2008. s. 1-7.

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

TY - GEN

T1 - Fusion of Range and Intensity Information for View Invariant Gesture Recognition

AU - Holte, Michael Boelstoft

AU - Moeslund, Thomas B.

AU - Fihl, Preben

PY - 2008

Y1 - 2008

N2 - This paper presents a system for view invariant gesture recognition. The approach is based on 3D data from a CSEM SwissRanger SR-2 camera. This camera produces both a depth map as well as an intensity image of a scene. Since the two information types are aligned, we can use the intensity image to define a region of interest for the relevant 3D data. This data fusion improves the quality of the range data and hence results in better recognition. The gesture recognition is based on finding motion primitives in the 3D data. The primitives are represented compactly and view invariant using harmonic shape context. A probabilistic Edit Distance classifier is applied to identify which gesture best describes a string of primitives. The approach is trained on data from one viewpoint and tested on data from a different viewpoint. The recognition rate is 92.9% which is similar to the recognition rate when training and testing on gestures from the same viewpoint, hence the approach is indeed view invariant.

AB - This paper presents a system for view invariant gesture recognition. The approach is based on 3D data from a CSEM SwissRanger SR-2 camera. This camera produces both a depth map as well as an intensity image of a scene. Since the two information types are aligned, we can use the intensity image to define a region of interest for the relevant 3D data. This data fusion improves the quality of the range data and hence results in better recognition. The gesture recognition is based on finding motion primitives in the 3D data. The primitives are represented compactly and view invariant using harmonic shape context. A probabilistic Edit Distance classifier is applied to identify which gesture best describes a string of primitives. The approach is trained on data from one viewpoint and tested on data from a different viewpoint. The recognition rate is 92.9% which is similar to the recognition rate when training and testing on gestures from the same viewpoint, hence the approach is indeed view invariant.

M3 - Article in proceeding

SN - 9781424423392

SP - 1

EP - 7

BT - Computer Vision and Pattern Recognition Workshops, 2008

PB - Electrical Engineering/Electronics, Computer, Communications and Information Technology Association

ER -

Holte MB, Moeslund TB, Fihl P. Fusion of Range and Intensity Information for View Invariant Gesture Recognition. I Computer Vision and Pattern Recognition Workshops, 2008. Electrical Engineering/Electronics, Computer, Communications and Information Technology Association. 2008. s. 1-7