“…It has been made publicly available to help and encourage more research in Bangla Sign Language Interpretation so that hearing and speech-impaired people can benefit from it. Most of the previous work on the BDSL dataset deals with around 30–38 labels [2] , [3] , [4] , [5] , but this dataset deals with 49 labels that have not been applied anywhere else. Two datasets have been produced from these images after acquiring 14,745 images from a few individuals.…”