Grasp type revisited

A modern perspective on a classical feature for vision

Yezhou Yang, Cornelia Fermüller, Yi Li, Yiannis Aloimonos

Research output: Chapter in Book/Report/Conference proceedingConference contribution

20 Citations (Scopus)

Abstract

The grasp type provides crucial information about human action. However, recognizing the grasp type from unconstrained scenes is challenging because of the large variations in appearance, occlusions and geometric distortions. In this paper, first we present a convolutional neural network to classify functional hand grasp types. Experiments on a public static scene hand data set validate good performance of the presented method. Then we present two applications utilizing grasp type classification: (a) inference of human action intention and (b) fine level manipulation action segmentation. Experiments on both tasks demonstrate the usefulness of grasp type as a cognitive feature for computer vision. This study shows that the grasp type is a powerful symbolic representation for action understanding, and thus opens new avenues for future research.

Original languageEnglish (US)
Title of host publicationIEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015
PublisherIEEE Computer Society
Pages400-408
Number of pages9
Volume07-12-June-2015
ISBN (Electronic)9781467369640
DOIs
StatePublished - Oct 14 2015
Externally publishedYes
EventIEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015 - Boston, United States
Duration: Jun 7 2015Jun 12 2015

Other

OtherIEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015
CountryUnited States
CityBoston
Period6/7/156/12/15

Fingerprint

Computer vision
Experiments
Neural networks

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition

Cite this

Yang, Y., Fermüller, C., Li, Y., & Aloimonos, Y. (2015). Grasp type revisited: A modern perspective on a classical feature for vision. In IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015 (Vol. 07-12-June-2015, pp. 400-408). [7298637] IEEE Computer Society. https://doi.org/10.1109/CVPR.2015.7298637

Grasp type revisited : A modern perspective on a classical feature for vision. / Yang, Yezhou; Fermüller, Cornelia; Li, Yi; Aloimonos, Yiannis.

IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015. Vol. 07-12-June-2015 IEEE Computer Society, 2015. p. 400-408 7298637.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Yang, Y, Fermüller, C, Li, Y & Aloimonos, Y 2015, Grasp type revisited: A modern perspective on a classical feature for vision. in IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015. vol. 07-12-June-2015, 7298637, IEEE Computer Society, pp. 400-408, IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015, Boston, United States, 6/7/15. https://doi.org/10.1109/CVPR.2015.7298637
Yang Y, Fermüller C, Li Y, Aloimonos Y. Grasp type revisited: A modern perspective on a classical feature for vision. In IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015. Vol. 07-12-June-2015. IEEE Computer Society. 2015. p. 400-408. 7298637 https://doi.org/10.1109/CVPR.2015.7298637
Yang, Yezhou ; Fermüller, Cornelia ; Li, Yi ; Aloimonos, Yiannis. / Grasp type revisited : A modern perspective on a classical feature for vision. IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015. Vol. 07-12-June-2015 IEEE Computer Society, 2015. pp. 400-408
@inproceedings{5b31438cda61465cb549395cb51b94c4,
title = "Grasp type revisited: A modern perspective on a classical feature for vision",
abstract = "The grasp type provides crucial information about human action. However, recognizing the grasp type from unconstrained scenes is challenging because of the large variations in appearance, occlusions and geometric distortions. In this paper, first we present a convolutional neural network to classify functional hand grasp types. Experiments on a public static scene hand data set validate good performance of the presented method. Then we present two applications utilizing grasp type classification: (a) inference of human action intention and (b) fine level manipulation action segmentation. Experiments on both tasks demonstrate the usefulness of grasp type as a cognitive feature for computer vision. This study shows that the grasp type is a powerful symbolic representation for action understanding, and thus opens new avenues for future research.",
author = "Yezhou Yang and Cornelia Ferm{\"u}ller and Yi Li and Yiannis Aloimonos",
year = "2015",
month = "10",
day = "14",
doi = "10.1109/CVPR.2015.7298637",
language = "English (US)",
volume = "07-12-June-2015",
pages = "400--408",
booktitle = "IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015",
publisher = "IEEE Computer Society",
address = "United States",

}

TY - GEN

T1 - Grasp type revisited

T2 - A modern perspective on a classical feature for vision

AU - Yang, Yezhou

AU - Fermüller, Cornelia

AU - Li, Yi

AU - Aloimonos, Yiannis

PY - 2015/10/14

Y1 - 2015/10/14

N2 - The grasp type provides crucial information about human action. However, recognizing the grasp type from unconstrained scenes is challenging because of the large variations in appearance, occlusions and geometric distortions. In this paper, first we present a convolutional neural network to classify functional hand grasp types. Experiments on a public static scene hand data set validate good performance of the presented method. Then we present two applications utilizing grasp type classification: (a) inference of human action intention and (b) fine level manipulation action segmentation. Experiments on both tasks demonstrate the usefulness of grasp type as a cognitive feature for computer vision. This study shows that the grasp type is a powerful symbolic representation for action understanding, and thus opens new avenues for future research.

AB - The grasp type provides crucial information about human action. However, recognizing the grasp type from unconstrained scenes is challenging because of the large variations in appearance, occlusions and geometric distortions. In this paper, first we present a convolutional neural network to classify functional hand grasp types. Experiments on a public static scene hand data set validate good performance of the presented method. Then we present two applications utilizing grasp type classification: (a) inference of human action intention and (b) fine level manipulation action segmentation. Experiments on both tasks demonstrate the usefulness of grasp type as a cognitive feature for computer vision. This study shows that the grasp type is a powerful symbolic representation for action understanding, and thus opens new avenues for future research.

UR - http://www.scopus.com/inward/record.url?scp=84959207244&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84959207244&partnerID=8YFLogxK

U2 - 10.1109/CVPR.2015.7298637

DO - 10.1109/CVPR.2015.7298637

M3 - Conference contribution

VL - 07-12-June-2015

SP - 400

EP - 408

BT - IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015

PB - IEEE Computer Society

ER -