Sign language spotting based on semi-Markov Conditional Random Field

Seong S. Cho, Hee Deok Yang, Seong Whan Lee

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

2 Citations (Scopus)

Abstract

Sign language spotting is the task of detecting the start and end points of signs from continuous data and recognizing the detected signs in the predefined vocabulary. The difficulty with sign language spotting is that instances of signs vary in terms of both motion and shape. Moreover, signs have variable motion in terms of both trajectory and length. Especially, variable sign lengths result in problems with spotting signs in a video sequence, because short signs involve less information and fewer changes than long signs. In this paper, we propose a method for spotting variable lengths signs based on semi-CRF (semi-Markov Conditional Random Field). We performed experiments with ASL (American Sign Language) and KSL (Korean Sign Language) datasets of continuous sign sentences to demonstrate the efficiency of the proposed method. Experimental results showed that the proposed method outperforms both HMM and CRF.

Original languageEnglish
Title of host publication2009 Workshop on Applications of Computer Vision, WACV 2009
DOIs
Publication statusPublished - 2009 Dec 1
Event2009 Workshop on Applications of Computer Vision, WACV 2009 - Snowbird, UT, United States
Duration: 2009 Dec 72009 Dec 8

Other

Other2009 Workshop on Applications of Computer Vision, WACV 2009
CountryUnited States
CitySnowbird, UT
Period09/12/709/12/8

Fingerprint

Trajectories
Experiments

ASJC Scopus subject areas

  • Computer Science Applications
  • Computer Vision and Pattern Recognition

Cite this

Cho, S. S., Yang, H. D., & Lee, S. W. (2009). Sign language spotting based on semi-Markov Conditional Random Field. In 2009 Workshop on Applications of Computer Vision, WACV 2009 [5403109] https://doi.org/10.1109/WACV.2009.5403109

Sign language spotting based on semi-Markov Conditional Random Field. / Cho, Seong S.; Yang, Hee Deok; Lee, Seong Whan.

2009 Workshop on Applications of Computer Vision, WACV 2009. 2009. 5403109.

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

Cho, SS, Yang, HD & Lee, SW 2009, Sign language spotting based on semi-Markov Conditional Random Field. in 2009 Workshop on Applications of Computer Vision, WACV 2009., 5403109, 2009 Workshop on Applications of Computer Vision, WACV 2009, Snowbird, UT, United States, 09/12/7. https://doi.org/10.1109/WACV.2009.5403109
Cho SS, Yang HD, Lee SW. Sign language spotting based on semi-Markov Conditional Random Field. In 2009 Workshop on Applications of Computer Vision, WACV 2009. 2009. 5403109 https://doi.org/10.1109/WACV.2009.5403109
Cho, Seong S. ; Yang, Hee Deok ; Lee, Seong Whan. / Sign language spotting based on semi-Markov Conditional Random Field. 2009 Workshop on Applications of Computer Vision, WACV 2009. 2009.
@inproceedings{12036d2b00794b3496dd80cbba300803,
title = "Sign language spotting based on semi-Markov Conditional Random Field",
abstract = "Sign language spotting is the task of detecting the start and end points of signs from continuous data and recognizing the detected signs in the predefined vocabulary. The difficulty with sign language spotting is that instances of signs vary in terms of both motion and shape. Moreover, signs have variable motion in terms of both trajectory and length. Especially, variable sign lengths result in problems with spotting signs in a video sequence, because short signs involve less information and fewer changes than long signs. In this paper, we propose a method for spotting variable lengths signs based on semi-CRF (semi-Markov Conditional Random Field). We performed experiments with ASL (American Sign Language) and KSL (Korean Sign Language) datasets of continuous sign sentences to demonstrate the efficiency of the proposed method. Experimental results showed that the proposed method outperforms both HMM and CRF.",
author = "Cho, {Seong S.} and Yang, {Hee Deok} and Lee, {Seong Whan}",
year = "2009",
month = "12",
day = "1",
doi = "10.1109/WACV.2009.5403109",
language = "English",
isbn = "9781424454976",
booktitle = "2009 Workshop on Applications of Computer Vision, WACV 2009",

}

TY - GEN

T1 - Sign language spotting based on semi-Markov Conditional Random Field

AU - Cho, Seong S.

AU - Yang, Hee Deok

AU - Lee, Seong Whan

PY - 2009/12/1

Y1 - 2009/12/1

N2 - Sign language spotting is the task of detecting the start and end points of signs from continuous data and recognizing the detected signs in the predefined vocabulary. The difficulty with sign language spotting is that instances of signs vary in terms of both motion and shape. Moreover, signs have variable motion in terms of both trajectory and length. Especially, variable sign lengths result in problems with spotting signs in a video sequence, because short signs involve less information and fewer changes than long signs. In this paper, we propose a method for spotting variable lengths signs based on semi-CRF (semi-Markov Conditional Random Field). We performed experiments with ASL (American Sign Language) and KSL (Korean Sign Language) datasets of continuous sign sentences to demonstrate the efficiency of the proposed method. Experimental results showed that the proposed method outperforms both HMM and CRF.

AB - Sign language spotting is the task of detecting the start and end points of signs from continuous data and recognizing the detected signs in the predefined vocabulary. The difficulty with sign language spotting is that instances of signs vary in terms of both motion and shape. Moreover, signs have variable motion in terms of both trajectory and length. Especially, variable sign lengths result in problems with spotting signs in a video sequence, because short signs involve less information and fewer changes than long signs. In this paper, we propose a method for spotting variable lengths signs based on semi-CRF (semi-Markov Conditional Random Field). We performed experiments with ASL (American Sign Language) and KSL (Korean Sign Language) datasets of continuous sign sentences to demonstrate the efficiency of the proposed method. Experimental results showed that the proposed method outperforms both HMM and CRF.

UR - http://www.scopus.com/inward/record.url?scp=77951145837&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77951145837&partnerID=8YFLogxK

U2 - 10.1109/WACV.2009.5403109

DO - 10.1109/WACV.2009.5403109

M3 - Conference contribution

SN - 9781424454976

BT - 2009 Workshop on Applications of Computer Vision, WACV 2009

ER -