Exploring Sign Language in Complex Background: Techniques, Datasets and Challenges

The Deaf community plays an important and vibrant role in society with sign language is used as primary means of communication. Sign language often remains confined within their community due to a lack of understanding among those without hearing impairments. SLR (Sign Language Recognition) solves this problem by enabling better communication and understanding of sign language among common people. This paper represents critical and systematic review of 71 articles for Sign recognition. In this paper we review how SLR makes significant strides, particularly through the development of innovative algorithm-driven techniques in both traditional and deep learning approaches. This review provides an overview of these advancements, focusing on main methodologies, including traditional methods and deep learning. It also examines available sign language datasets, identifies the challenges facing SLR with respect to different complexity by analysing and comparing the fundamental principles of various SLR techniques corresponding to complex patterns.