Real-Time Recognition and Translation of Kinyarwanda Sign Language into Kinyarwanda Text (2023) Mediapipe NumPy array Hands and Pose extracted key points for 22 Sign Language

This research addresses the issue of real-time translation of sign language into text (focusing on Kinyarwanda Sign Language) concentrating on twenty-two common gestures in Kinyarwanda sign language. Through extensive exploration and evaluation of various machine learning algorithms, the study identifies the most effective approach for recognizing and translating these gestures. To validate the effectiveness of the developed system, real-world Kinyarwanda sign language video data is utilized for thorough training and testing.
The data set contains Hands and Pose Mediapipe extracted key points for the 22 sign language and one additional sign ("---" sign stands for not signing) saved in the NumPy array. It can be used to train the LSTM model for the classification of the 22 signs.

本研究围绕手语转文本的实时翻译问题展开，以卢旺达手语（Kinyarwanda Sign Language）为研究载体，聚焦该手语体系内的22种常用手势。通过对多种机器学习算法开展广泛探索与评估，本研究确定了用于识别并翻译上述手势的最优方案。为验证所研发系统的有效性，研究采用真实场景下的卢旺达手语视频数据开展全面的训练与测试工作。本数据集包含经Mediapipe手部与姿态模块提取得到的22种手语手势以及1种额外手势的关键点数据（"---"代表未做出手势），数据以NumPy数组格式存储。该数据集可用于训练长短期记忆（LSTM）模型，以完成22种手语手势的分类任务。