Survey of Hidden Markov Models (HMMs) for Sign Language Recognition (SLR)

This paper surveyed several significant papers on specific topics applying the Hidden Markov Models (HMMs) for Sign Language Recognition (SLR), divided into five main episodes: Classical HMMs, Extended HMMs, HMMs and Machine Learning, HMMs and Sensor Fusion, and HMMs and Big Data. This stringent survey would contribute significantly to advanced research on unification brain models such as neural networks, adaptive resonance theory, and confabulation theory. First, the HMM was introduced as one of the popular methods of performing SLR, and each episode of its development was expounded. In each episode, a main paper and several supporting papers were summarized. Next, HMM’s general methodology and the remarkable feature extraction methods from each episode should be discussed. The unique feature extraction methods and the Markov decision processes were highlighted. Third, each episode simulation or numerical results were presented, compared, and commented on. Finally, the paper concludes with the author’s takeaway and insight on each paper, especially the relation of HMMs with Recurrent Neural Networks, and provides a road map for future research. One innovation point of this paper is the relic of the suitable topology for both HMMs and recurrent neural networks for the SLR system.