SIBI Indonesian Sign Language Alphabet Image Dataset for Hand Gesture Recognition

Published: 4 June 2026| Version 1 | DOI: 10.17632/xdn2gj38xf.1
Contributors:
, Desi Fatkhi Azizah

Description

This dataset contains hand gesture images representing the alphabet of Sistem Isyarat Bahasa Indonesia (SIBI), an Indonesian sign language system used as a communication medium for deaf and hard-of-hearing communities in Indonesia. The dataset was developed to support research on image-based sign language recognition, hand gesture classification, object detection, and segmentation using machine learning and deep learning approaches.

Files

Steps to reproduce

The dataset was collected by capturing static hand gesture images representing the SIBI alphabet from A to Z. Images were acquired using a fixed camera position, adequate lighting, and a uniform plain background. The collected images were grouped into folders according to their alphabet labels. Unclear, duplicated, and incorrectly labeled images were removed before the dataset was used for preprocessing and sign language recognition experiments.

Institutions

Categories

Artificial Intelligence, Computer Vision

Licence