Skip navigation
Please use this identifier to cite or link to this item: http://repository.iitr.ac.in/handle/123456789/5696
Title: Dense motion analysis of German finger spellings
Authors: Tanwar V.K.
Buckchash H.
Raman, Balasubramanian
Bhargava R.
Published in: Multimedia Tools and Applications
Abstract: Finger spellings is a frequent form of sign-language communication used by hearing-impaired people. It is a powerful media for visual communication. Capturing sign-language modalities are often challenging due to quick compact motion of fingers. Other approaches use costly sensors such as leap-motion or specially designed gloves or color segmentation. These are not speaker invariant, and are less speaker-friendly as they do not allow free expression of gesture. This work focuses on the progressive finger motion aspects of the speaker by analyzing the motion of gestures. Hand is segmented with Gaussian Mixture Model. It boosts the calculation of dense trajectories to trace hand movement. A Bag-of-Word (BoW) representation is employed to capture motion features. Experiments are conducted on German finger spelling dataset to evaluate the performance of the proposed technique. The observed performance of the technique indicates the effectiveness of motion as a substantial modality for finger spelling recognition. © 2018, Springer Science+Business Media, LLC, part of Springer Nature.
Citation: Multimedia Tools and Applications (2019), 78(8): 9511-9536
URI: https://doi.org/10.1007/s11042-018-6533-0
http://repository.iitr.ac.in/handle/123456789/5696
Issue Date: 2019
Publisher: Springer New York LLC
Keywords: Dense trajectories
Finger spelling recognition
Gaussian mixture model
Sign language recognition
ISSN: 13807501
Author Scopus IDs: 57202340293
57196062798
23135470700
23063065000
Author Affiliations: Tanwar, V.K., Department of Mathematics, Indian Institute of Technology Roorkee, Roorkee, Uttarakhand, India
Buckchash, H., Department of Computer Science & Engineering, Indian Institute of Technology Roorkee, Roorkee, Uttarakhand, India
Raman, B., Department of Computer Science & Engineering, Indian Institute of Technology Roorkee, Roorkee, Uttarakhand, India
Bhargava, R., Department of Mathematics, Indian Institute of Technology Roorkee, Roorkee, Uttarakhand, India
Corresponding Author: Tanwar, V.K.; Department of Mathematics, Indian Institute of Technology RoorkeeIndia; email: vishu.dma2015@iitr.ac.in
Appears in Collections:Journal Publications [CS]

Files in This Item:
There are no files associated with this item.
Show full item record


Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.