Norwegian Sign Language

Published: 4 August 2023| Version 1 | DOI: 10.17632/3cfrj4vd4m.1
Contributors:
Seifedine Kadry, Benjamin Svendsen

Description

The dataset used in our research is a self-made NSL fingerspelling dataset containing 24 300 across 27 alphabets. Each alphabet has 600 images taken of a male where ½ of it is with basic white background and the other ½ is of complex background. Each alphabet also has 300 images taken of a female with a basic white background. We want this dataset to be available to the public so further research can be made in the field of NSL recognition.

Files

Institutions

  • Noroff Oslo

Categories

Image Acquisition, Image Classification, Sign Language

Licence