Extreme Learning Machine for Real Time Recognition of Brazilian Sign Language

Neto, Fernando M. de Paula; Cambuim, Lucas F. S.; Macieira, Rafael M.; Ludermir, Teresa B.; Zanchettin, Cleber; Barros, Edna

doi:10.1109/smc.2015.259

Cited by 9 publications

(2 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most literature on signal language recognition deal with the relative configuration of the hands [Escobedo-Cardenas and Camara-Chavez 2015], [Almeida et al 2014], [de Paula Neto et al 2015], [Pariwat and Seresangtakul 2017], [Uddin and Chowdhury 2016]. Most of these studies carry out recognition of the alphabet in their respective languages.…”

Section: Related Workmentioning

confidence: 99%

Facial Expression Analysis in Brazilian Sign Language for Sign Recognition

Guerra¹,

Rezende²,

Guimarães³

et al. 2018

Anais Do XV Encontro Nacional De Inteligência Artificial E Computacional (ENIAC 2018)

View full text Add to dashboard Cite

Sign language is one of the main forms of communication used by the deaf community. The language’s smallest unit, a “sign”, comprises a series of intricate manual and facial gestures. As opposed to speech recognition, sign language recognition (SLR) lags behind, presenting a multitude of open challenges because this language is visual-motor. This paper aims to explore two novel approaches in feature extraction of facial expressions in SLR, and to propose the use of Random Forest (RF) in Brazilian SLR as a scalable alternative to Support Vector Machines (SVM) and k-Nearest Neighbors (k-NN). Results show that RF’s performance is at least comparable to SVM’s and k-NN’s, and validate non-manual parameter recognition as a consistent step towards SLR.

show abstract

Section: Related Workmentioning

confidence: 99%

Facial Expression Analysis in Brazilian Sign Language for Sign Recognition

Guerra¹,

Rezende²,

Guimarães³

et al. 2018

Anais Do XV Encontro Nacional De Inteligência Artificial E Computacional (ENIAC 2018)

View full text Add to dashboard Cite

show abstract

“…In [8] the authors propose a gesture recognition system for the ASL utilizing the Wavelet transformation showing results of 97.4% accuracy. Aiming to improve the recognition time in [9] the authors propose an architecture of an embedded platform to recognize the Brazilian Sign Language (LIBRAS) based on FPGA and images obtained from a camera. This paper proposes a system for static gesture recognition of LIBRAS using Microsoft Kinect concomitantly with the eigenhands technique linked with the process of lighting normalization in the gesture images obtained through the Kinect.…”

Section: Introductionmentioning

confidence: 99%

Static Gestures Recognition for Brazilian Sign Language with Kinect and Eigenhands

Carneiro¹,

Santos²,

Barbosa³

et al. 2016

IJCA

View full text Add to dashboard Cite

The automatic recognition of sign languages to help hearing impaired people is an area that has been explored for quite some time. However, this is still a practical problem due to the complexity involved making it a big challenge. Environments with complex backgrounds and different lighting conditions, hinders the recognition of hand gestures and other expressions when performed. The use of devices that are developed to increase the man machine interaction for entertainment, such as the Microsoft Kinect, has been shown to be promising in gesture recognition due to the amount of resources that it provides in the development of applications involving movements. The ease in detecting users and regions of interest such as hands, greatly reduces the complexity in the process of capturing gestures made by a user in complex environments. Therefore, this paper proposes an application for users of the Brazilian Sign Language (Libras), which brings a main feature of voice generation based on static hand gestures. The application utilizes the Kinect technology concomitantly with the eigenhands technique linked with the process of lighting normalization in the gesture images. The preliminary results obtained through testing had an accuracy reaching up to 89% in the static gesture recognition of the language LIBRAS.

show abstract