Exploring MediaPipe optimization strategies for real-time sign language recognition

The present study meticulously investigates optimization strategies for real-time sign language recognition (SLR) employing the MediaPipe framework. We introduce an innovative multi-modal methodology, amalgamating four distinct Long Short-Term Memory (LSTM) models dedicated to processing skeletal c...

Full description

Saved in:
Bibliographic Details
Main Authors: Phuoc Thanh Nguyen, Thanh Hoang Nguyen, Ngoc Xuan Nguyen Hoang, Huynh Thanh Binh Phan, Hoang Son Hai Vu, Hieu Nhan Huynh
Format: Article
Language:English
Published: Can Tho University Publisher 2023-10-01
Series:CTU Journal of Innovation and Sustainable Development
Subjects:
Online Access:https://ctujs.ctu.edu.vn/index.php/ctujs/article/view/716
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849717596751921152
author Phuoc Thanh Nguyen
Thanh Hoang Nguyen
Ngoc Xuan Nguyen Hoang
Huynh Thanh Binh Phan
Hoang Son Hai Vu
Hieu Nhan Huynh
author_facet Phuoc Thanh Nguyen
Thanh Hoang Nguyen
Ngoc Xuan Nguyen Hoang
Huynh Thanh Binh Phan
Hoang Son Hai Vu
Hieu Nhan Huynh
author_sort Phuoc Thanh Nguyen
collection DOAJ
description The present study meticulously investigates optimization strategies for real-time sign language recognition (SLR) employing the MediaPipe framework. We introduce an innovative multi-modal methodology, amalgamating four distinct Long Short-Term Memory (LSTM) models dedicated to processing skeletal coordinates ascertained from the MediaPipe framework. Rigorous evaluations were executed on esteemed sign language datasets. Empirical findings underscore that the multi-modal approach significantly elevates the accuracy of the SLR model while preserving its real-time capabilities. In comparative analyses with prevalent MediaPipe-based models, our multi-modal strategy consistently manifested superior performance metrics. A distinguishing characteristic of this approach is its inherent adaptability, facilitating modifications within the LSTM layers, rendering it apt for a myriad of challenges and data typologies. Integrating the MediaPipe framework with real-time SLR markedly amplifies recognition precision, signifying a pivotal advancement in the discipline.
format Article
id doaj-art-c571062de5d9455e95c6d4646fb72f75
institution DOAJ
issn 2588-1418
2815-6412
language English
publishDate 2023-10-01
publisher Can Tho University Publisher
record_format Article
series CTU Journal of Innovation and Sustainable Development
spelling doaj-art-c571062de5d9455e95c6d4646fb72f752025-08-20T03:12:36ZengCan Tho University PublisherCTU Journal of Innovation and Sustainable Development2588-14182815-64122023-10-0115Special issue: ISDS10.22144/ctujoisd.2023.045Exploring MediaPipe optimization strategies for real-time sign language recognitionPhuoc Thanh Nguyen0Thanh Hoang NguyenNgoc Xuan Nguyen HoangHuynh Thanh Binh PhanHoang Son Hai Vu Hieu Nhan Huynh 0911427905 The present study meticulously investigates optimization strategies for real-time sign language recognition (SLR) employing the MediaPipe framework. We introduce an innovative multi-modal methodology, amalgamating four distinct Long Short-Term Memory (LSTM) models dedicated to processing skeletal coordinates ascertained from the MediaPipe framework. Rigorous evaluations were executed on esteemed sign language datasets. Empirical findings underscore that the multi-modal approach significantly elevates the accuracy of the SLR model while preserving its real-time capabilities. In comparative analyses with prevalent MediaPipe-based models, our multi-modal strategy consistently manifested superior performance metrics. A distinguishing characteristic of this approach is its inherent adaptability, facilitating modifications within the LSTM layers, rendering it apt for a myriad of challenges and data typologies. Integrating the MediaPipe framework with real-time SLR markedly amplifies recognition precision, signifying a pivotal advancement in the discipline. https://ctujs.ctu.edu.vn/index.php/ctujs/article/view/716LSTM, MediaPipe, How2Sign, Indian Sign Language, ISL
spellingShingle Phuoc Thanh Nguyen
Thanh Hoang Nguyen
Ngoc Xuan Nguyen Hoang
Huynh Thanh Binh Phan
Hoang Son Hai Vu
Hieu Nhan Huynh
Exploring MediaPipe optimization strategies for real-time sign language recognition
CTU Journal of Innovation and Sustainable Development
LSTM, MediaPipe, How2Sign, Indian Sign Language, ISL
title Exploring MediaPipe optimization strategies for real-time sign language recognition
title_full Exploring MediaPipe optimization strategies for real-time sign language recognition
title_fullStr Exploring MediaPipe optimization strategies for real-time sign language recognition
title_full_unstemmed Exploring MediaPipe optimization strategies for real-time sign language recognition
title_short Exploring MediaPipe optimization strategies for real-time sign language recognition
title_sort exploring mediapipe optimization strategies for real time sign language recognition
topic LSTM, MediaPipe, How2Sign, Indian Sign Language, ISL
url https://ctujs.ctu.edu.vn/index.php/ctujs/article/view/716
work_keys_str_mv AT phuocthanhnguyen exploringmediapipeoptimizationstrategiesforrealtimesignlanguagerecognition
AT thanhhoangnguyen exploringmediapipeoptimizationstrategiesforrealtimesignlanguagerecognition
AT ngocxuannguyenhoang exploringmediapipeoptimizationstrategiesforrealtimesignlanguagerecognition
AT huynhthanhbinhphan exploringmediapipeoptimizationstrategiesforrealtimesignlanguagerecognition
AT hoangsonhaivu exploringmediapipeoptimizationstrategiesforrealtimesignlanguagerecognition
AT hieunhanhuynh exploringmediapipeoptimizationstrategiesforrealtimesignlanguagerecognition