Hand gesture recognition with 3D convolutional neural networks
P. Molchanov, S. Gupta, K. Kim, and J. Kautz. The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, page 1-7. IEEE, (September 2015)
Abstract
Touchless hand gesture recognition systems are becoming important in automotive user interfaces as they improve safety and comfort. Various computer vision algorithms have employed color and depth cameras for hand gesture recognition, but robust classification of gestures from different subjects performed under widely varying lighting conditions is still challenging. We propose an algorithm for drivers’ hand gesture recognition from challenging depth and intensity data using 3D convolutional neural networks. Our solution combines information from multiple spatial scales for the final prediction. It also employs spatiotemporal data augmentation for more effective training and to reduce potential overfitting. Our method achieves a correct classification rate of 77.5% on the VIVA challenge dataset.
%0 Conference Paper
%1 molchanov2015gesture
%A Molchanov, Pavlo
%A Gupta, Shalini
%A Kim, Kihwan
%A Kautz, Jan
%B The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops
%D 2015
%I IEEE
%K 3dcnn gesture-recognition real sign-language-recognition support-vector-machine svm
%P 1-7
%T Hand gesture recognition with 3D convolutional neural networks
%U https://www.cv-foundation.org/openaccess/content_cvpr_workshops_2015/W15/html/Molchanov_Hand_Gesture_Recognition_2015_CVPR_paper.html
%X Touchless hand gesture recognition systems are becoming important in automotive user interfaces as they improve safety and comfort. Various computer vision algorithms have employed color and depth cameras for hand gesture recognition, but robust classification of gestures from different subjects performed under widely varying lighting conditions is still challenging. We propose an algorithm for drivers’ hand gesture recognition from challenging depth and intensity data using 3D convolutional neural networks. Our solution combines information from multiple spatial scales for the final prediction. It also employs spatiotemporal data augmentation for more effective training and to reduce potential overfitting. Our method achieves a correct classification rate of 77.5% on the VIVA challenge dataset.
%@ 978-1-4673-6759-2
@inproceedings{molchanov2015gesture,
abstract = {Touchless hand gesture recognition systems are becoming important in automotive user interfaces as they improve safety and comfort. Various computer vision algorithms have employed color and depth cameras for hand gesture recognition, but robust classification of gestures from different subjects performed under widely varying lighting conditions is still challenging. We propose an algorithm for drivers’ hand gesture recognition from challenging depth and intensity data using 3D convolutional neural networks. Our solution combines information from multiple spatial scales for the final prediction. It also employs spatiotemporal data augmentation for more effective training and to reduce potential overfitting. Our method achieves a correct classification rate of 77.5% on the VIVA challenge dataset.},
added-at = {2019-11-14T18:37:27.000+0100},
author = {Molchanov, Pavlo and Gupta, Shalini and Kim, Kihwan and Kautz, Jan},
biburl = {https://www.bibsonomy.org/bibtex/2bde3f59ed9fd7818df39826bf9623fea/jpmor},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
crossref = {conf/cvpr/2015w},
ee = {http://doi.ieeecomputersociety.org/10.1109/CVPRW.2015.7301342},
institution = {NVIDIA},
interhash = {d7c191979b48e5e4f1c1544512fee944},
intrahash = {bde3f59ed9fd7818df39826bf9623fea},
isbn = {978-1-4673-6759-2},
keywords = {3dcnn gesture-recognition real sign-language-recognition support-vector-machine svm},
language = {English},
month = {09},
pages = {1-7},
publisher = {IEEE},
timestamp = {2020-10-07T13:36:50.000+0200},
title = {Hand gesture recognition with 3D convolutional neural networks},
url = {https://www.cv-foundation.org/openaccess/content_cvpr_workshops_2015/W15/html/Molchanov_Hand_Gesture_Recognition_2015_CVPR_paper.html},
year = 2015
}