Classification of Czech Sign Language Alphabet Diacritics via LSTM

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26210%2F22%3APU147803" target="_blank" >RIV/00216305:26210/22:PU147803 - isvavai.cz</a>

Výsledek na webu

<a href="https://ieeexplore.ieee.org/document/9983436/" target="_blank" >https://ieeexplore.ieee.org/document/9983436/</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/ME54704.2022.9983436" target="_blank" >10.1109/ME54704.2022.9983436</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Classification of Czech Sign Language Alphabet Diacritics via LSTM

Popis výsledku v původním jazyce

The paper presents the classification of image sequences of the single-handed Czech sign language alphabet diacritics. Since diacritics are expressed by the motion of the hand, the classification is performed by the Long Short-Term Memory recurrent neural network. Annotation of the dataset is done by the MediaPipe framework, and the neural network is constructed with the TensorFlow computational library. The paper describes the proposed method's flow, data acquisition, preprocessing, and training. Obtained results consist of the validation dataset's classification success rate and testing on whole signed words and sentences. The overall success rate was around 88%.

Název v anglickém jazyce

Classification of Czech Sign Language Alphabet Diacritics via LSTM

Popis výsledku anglicky

The paper presents the classification of image sequences of the single-handed Czech sign language alphabet diacritics. Since diacritics are expressed by the motion of the hand, the classification is performed by the Long Short-Term Memory recurrent neural network. Annotation of the dataset is done by the MediaPipe framework, and the neural network is constructed with the TensorFlow computational library. The paper describes the proposed method's flow, data acquisition, preprocessing, and training. Obtained results consist of the validation dataset's classification success rate and testing on whole signed words and sentences. The overall success rate was around 88%.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2022

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

2022 20th International Conference on Mechatronics - Mechatronika (ME)

ISBN

978-1-6654-1039-7

ISSN

—

e-ISSN

—

Počet stran výsledku

5

Strana od-do

178-182

Název nakladatele

Institute of Electrical and Electronics Engineers Inc.

Místo vydání

Pilsen

Místo konání akce

Pilsen

Datum konání akce

7. 12. 2022

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—