ISLR101
收藏资源简介:
ISLR101是一个公开的伊朗手语词汇级识别数据集,由Sharif University of Technology的Social and Cognitive Robotics Lab.创建。该数据集包含4614个视频,涵盖101个不同的手语词汇,由10位手语者表演。每个词汇至少由8位不同的手语者表演,这有利于提高训练出的手语识别模型的泛化能力。该数据集适用于词级手语识别研究,可促进手语识别技术的发展。
ISLR101 is a publicly available lexicon-level Iranian Sign Language recognition dataset, developed by the Social and Cognitive Robotics Lab at Sharif University of Technology. This dataset comprises 4614 videos covering 101 distinct sign language vocabulary items, performed by 10 signers. Each vocabulary item is performed by at least 8 different signers, which helps improve the generalization capability of trained sign language recognition models. This dataset is suitable for word-level sign language recognition research and can promote the development of sign language recognition technologies.

- 1ISLR101: an Iranian Word-Level Sign Language Recognition Dataset Sharif University of Technology · 2025年



