Content area

Abstract

When developing a continuous sign language recognition (CSLR) system, a significant challenge lies in processing the vast number of video frames, which demands extensive time and computational resources during both the training and prediction phases. To address this, we propose an efficient and scalable methodology that integrates cluster-based key frame extraction with a VOGUE-based recognition model designed for continuous gestures. The key frame extraction strategy clusters visually similar frames to reduce redundancy while preserving only those with high semantic relevance. To further enhance recognition accuracy, we introduce the Key Curvature Maximum Point (KCMP) technique, which identifies pivotal motion points and captures essential hand trajectory changes inherent to sign language. These refined frames are subsequently used to train a VOGUE-based model that encodes spatial and temporal strokes dynamics, followed by probability distribution modeling for robust prediction. The proposed approach was evaluated using a custom-built Tamil Sign Language dataset. Performance was compared against several established baseline methods, including Dynamic Time Warping (DTW), Hidden Markov Models (HMM), and multiple Conditional Random Field (CRF) variants, as well as the VOM model. The system achieved a recognition accuracy of 86.78% and a sign error rate of 5.3%. A paired t-test confirmed that the improvements over baseline models were statistically significant (p < 0.05). These results demonstrate that the proposed framework provides improved efficiency and competitive accuracy, offering a promising solution for real-time CSLR applications, particularly in low-resource regional sign languages.

Details

Business indexing term
Title
VOGUE-Based Approach for Segmenting Movement Epenthesis in Continuous Sign Language Recognition
Author
Publication title
Volume
30
Issue
11
Pages
2949-2959
Number of pages
12
Publication year
2025
Publication date
Nov 2025
Publisher
International Information and Engineering Technology Association (IIETA)
Place of publication
Edmonton
Country of publication
Canada
ISSN
16331311
e-ISSN
21167125
Source type
Scholarly Journal
Language of publication
French
Document type
Journal Article
Publication history
 
 
Online publication date
2025-11-30
Milestone dates
2025-11-18 (Accepted); 2025-11-10 (Revised); 2025-09-01 (Received)
Publication history
 
 
   First posting date
30 Nov 2025
ProQuest document ID
3293380337
Document URL
https://www.proquest.com/scholarly-journals/vogue-based-approach-segmenting-movement/docview/3293380337/se-2?accountid=208611
Copyright
© 2025. This work is published under https://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Last updated
2026-01-15
Database
ProQuest One Academic