NITJ-ISLC-CB sign language character dataset

This repository contains a representative subset of the NITJ-ISLC-CB sign language dataset collected from child signers at a school for deaf children. The dataset comprises 36 Indian Sign Language gesture categories captured under naturalistic conditions with complex real-world backgrounds including cluttered indoor scenes and varied lighting conditions. All images in this publicly released subset have been anonymised through automated face removal to protect the identity of minor participants. No personally identifiable information is present in any image. The complete dataset is available upon reasonable request for academic research purposes. Please contact the corresponding author at [e-mail]. Citation : Bahia, N. (2026). NITJ-ISLC-CB sign language character dataset (Version Version 1) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.21713512