Sign language identification and recognition: A comparative study

Abstract Sign Language (SL) is the main language for handicapped and disabled people. Each country has its own SL that different from other countries. sign in a represented with variant hand gestures, body movements, facial expressions. Researchers this field aim to remove any obstacles prevent communication deaf people by replacing all device-based techniques vision-based using Artificial Intelligence (AI) Deep Learning. This article highlights two processing tasks: Recognition (SLR) Identification (SLID). The latter task targeted identify signer language, while former aimed translate conversation into tokens (signs). addresses most common datasets used literature tasks (static dynamic are collected corpora) contents including numerical, alphabets, words, sentences SLs. It also discusses devices required build these datasets, as well preprocessing steps applied before training testing. compares approaches on datasets. both data-gloves-based approaches, aiming analyze focus methods such hybrid deep learning algorithms. Furthermore, presents graphical depiction tabular representation of various SLR approaches.