Beyond the Laboratory Illusion: A PRISMA-Guided Systematic Review of Arabic Sign Language Processing (2015–2025)

Muna G. Abd Alkreem; Baidaa  Mutasher Rashed; Rasha B. Yousif

doi:10.71229/5ehtp096

Authors

Muna G. Abd Alkreem Department of Computer Science, College of Computer Science and Mathematics,University of Thi-Qar,Iraq
Baidaa Mutasher Rashed Department of Computers, College of Computer Science & Mathematics, University of Thi-Qar, Nasiriyah , Thi-Qar, Iraq
Rasha B. Yousif Department of Computer Science, College of Computer Science and Mathematics, University of Thi Qar, Iraq

DOI:

https://doi.org/10.71229/5ehtp096

Keywords:

Keywords: Arabic Sign Language Processing; Systematic Review; Computational Tasks; Model Architectures; Architectural Hybr

Abstract

Arabic Sign Language (ArSL) processing research has expanded rapidly over the last 10 years. However, a gap exists between innovative algorithms introduced in research and their implementation in society. To help close this gap, the current research offers a systematic review of ArSL processing for the years 2015-2025. Within the PRISMA framework, we provided a multi-dimensional analysis of 33 empirical papers and mapped the main research obstacles, as well as the computational methods and model architectures. This analysis showed a significant imbalance in data availability (32.3%) and a lack of ability to generalize to new domains or tasks (29.0%). The reported accuracy in these studies can be considered "illusions" because those results were likely achieved in laboratory environments. Overall, the available literature described three main eras: Foundation & Exploration, Deep & Dynamic Shift, and Generalization & Generation. Architecturally, the intricacy of spatiotemporal dynamics has called for 'architectural hybridization' (25.0%), along with operational bifurcation toward edge-efficient lightweight models. Furthermore, a distinct paradigm shift is occurring, abandoning intrusive sensors for non-intrusive natural vision, accompanied by a shift from visual decoding of a solitary nature to interactive linguistic generation [1]. The review finally states that future progress in ArSL processing will not only rely on enhancing the algorithms; an immediate shift is required toward data-driven technologies, and embracing the self-supervised paradigm to overcome the annotation bottleneck, and the co-design approach with the Deaf community to ensure the technological deployment is fair and useful.

References

[1] O. Koller, N. C. Camgoz, H. Ney, and R. Bowden, “Weakly Supervised Learning with Multi-Stream CNN-LSTM-HMMs to Discover Sequential Parallelism in Sign Language Videos,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 42, no. 9, pp. 2306–2320, 2020, doi: 10.1109/TPAMI.2019.2911077.

[2] M. A. Abdel-Fattah, “Arabic sign language: A perspective,” J. Deaf Stud. Deaf Educ., vol. 10, no. 2, pp. 212–221, 2005, doi: 10.1093/deafed/eni007.

[3] D. S. Almubayei, “Sign language choice and policy among the signing community in Kuwait,” Dig. Middle East Stud., vol. 33, no. 2, pp. 166–183, 2024, [Online]. Available: /doi/pdf/10.1111/dome.12316

[4] O. Koller, H. Ney, and R. Bowden, “Deep Hand: How to Train a CNN on 1 Million Hand Images When Your Data is Continuous and Weakly Labelled,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 2016-Decem, pp. 3793–3802, 2016, doi: 10.1109/CVPR.2016.412.

[5] B. Hisham and A. Hamouda, “Supervised learning classifiers for Arabic gestures recognition using Kinect V2,” SN Appl. Sci., vol. 1, no. 7, pp. 1–21, 2019, doi: 10.1007/s42452-019-0771-2.

[6] M. A. Almasre and H. Al-Nuaim, “A comparison of Arabic sign language dynamic gesture recognition models,” Heliyon, vol. 6, no. 3, p. e03554, 2020, doi: 10.1016/j.heliyon.2020.e03554.

[7] M. Al-Qurishi, T. Khalid, and R. Souissi, “Deep Learning for Sign Language Recognition: Current Techniques, Benchmarks, and Open Issues,” IEEE Access, vol. 9, pp. 126917–126951, 2021, doi: 10.1109/ACCESS.2021.3110912.

[8] B. Hisham and A. Hamouda, “Arabic sign language recognition using Ada-Boosting based on a leap motion controller,” Int. J. Inf. Technol., vol. 13, no. 3, pp. 1221–1234, 2021, doi: 10.1007/s41870-020-00518-5.

[9] S. Rouabhi, R. Tlemsani, and N. Neggaz, “Real-time mobile application for Arabic sign alphabet recognition using pre-trained CNN,” Soft Comput., vol. 28, no. 21, pp. 12991–13008, 2024, doi: 10.1007/s00500-024-10305-0.

[10] R. Rastgoo, K. Kiani, and S. Escalera, “Sign Language Recognition: A Deep Survey,” Expert Syst. Appl., vol. 164, p. 113794, 2021, [Online]. Available: https://www.sciencedirect.com/science/article/abs/pii/S095741742030614X

[11] A. Duarte et al., “How2Sign: A Large-scale Multimodal Dataset for Continuous American Sign Language Gloss Annotation,” Cvpr, pp. 1–10, 2021, [Online]. Available: http://how2sign.github.io/

[12] A. S. Al-Shamayleh, R. Ahmad, N. Jomhari, and M. A. M. Abushariah, “Automatic Arabic sign language recognition: A review, taxonomy, open challenges, research roadmap and future directions‏,” Malaysian J. Comput. Sci., vol. 33, no. 4, pp. 306–343, 2020.

[13] A. Alayed, “Machine Learning and Deep Learning Approaches for Arabic Sign Language Recognition: A Decade Systematic Literature Review,” Sensors, vol. 24, no. 23, 2024, doi: 10.3390/s24237798.

[14] M. J. Page et al., “The PRISMA 2020 statement: An updated guideline for reporting systematic reviews,” Bmj, vol. 372, 2021, doi: 10.1136/bmj.n71.

[15] V. Braun and V. Clarke, “Qualitative Research in Psychology Using thematic analysis in psychology Using thematic analysis in psychology,” Qual. Res. Psychol., vol. 3, no. 2, pp. 77–101, 2006, [Online]. Available: http://www.tandfonline.com/action/journalInformation?journalCode=uqrp20%5Cnhttp://www.tandfonline.com/action/journalInformation?journalCode=uqrp20

[16] M. Alsulaiman et al., “Facilitating the communication with deaf people: Building a largest Saudi sign language dataset,” J. King Saud Univ. - Comput. Inf. Sci., vol. 35, no. 8, p. 101642, 2023, doi: 10.1016/j.jksuci.2023.101642.

[17] S. Abbas, D. Alahmadi, and H. Al-Barhamtoshy, “Establishing a multimodal dataset for Arabic Sign Language (ArSL) production,” J. King Saud Univ. - Comput. Inf. Sci., vol. 36, no. 8, p. 102165, 2024, doi: 10.1016/j.jksuci.2024.102165.

[18] A. H. AlQallaf, “Development of a web-based Unified Arabic/American sign language bilingual dictionary,” J. Eng. Res., vol. 6, no. 2, pp. 84–102, 2018, doi: 10.1016/s2307-1877(25)00811-9.

[19] N. B. Ibrahim, M. M. Selim, and H. H. Zayed, “An Automatic Arabic Sign Language Recognition System (ArSLRS),” J. King Saud Univ. - Comput. Inf. Sci., vol. 30, no. 4, pp. 470–477, 2018, doi: 10.1016/j.jksuci.2017.09.007.

[20] S. M. Elatawy, D. M. Hawa, A. A. Ewees, and A. M. Saad, “Recognition system for alphabet Arabic sign language using neutrosophic and fuzzy c-means,” Educ. Inf. Technol., vol. 25, no. 6, pp. 5601–5616, 2020, doi: 10.1007/s10639-020-10184-6.

[21] A. F. Alnabih and A. Y. Maghari, “Arabic sign language letters recognition using Vision Transformer,” Multimed. Tools Appl. 2024 8334, vol. 83, no. 34, pp. 81725–81739, 2024, [Online]. Available: https://link.springer.com/article/10.1007/s11042-024-18681-3

[22] W. Abdul et al., “Intelligent real-time Arabic sign language classification using attention-based inception and BiLSTM,” Comput. Electr. Eng., vol. 95, no. April, p. 107395, 2021, doi: 10.1016/j.compeleceng.2021.107395.

[23] A. A. Mohamed, A. Al-Saleh, S. K. Sharma, and G. Tejani, “Attention-based hybrid deep learning model with CSFOA optimization and G-TverskyUNet3+ for Arabic sign language recognition,” Sci. Rep., vol. 15, no. 1, pp. 1–33, 2025, doi: 10.1038/s41598-025-03560-0.

[24] A. Boukdir, M. Benaddy, O. El Meslouhi, M. Kardouchi, and M. Akhloufi, “Character-level arabic text generation from sign language video using encoder–decoder model,” Displays, vol. 76, no. July 2022, 2023, doi: 10.1016/j.displa.2022.102340.

[25] M. Brour and A. Benabbou, “ATLASLang MTS 1: Arabic Text Language into Arabic Sign Language Machine Translation System,” Procedia Comput. Sci., vol. 148, no. Icds 2018, pp. 236–245, 2019, doi: 10.1016/j.procs.2019.01.066.

[26] A. Abdalla, A. Alsereidi, N. Alyammahi, F. B. Qehaizel, H. A. Ignatious, and H. El-Sayed, “An Innovative Arabic Text Sign Language Translator,” Procedia Comput. Sci., vol. 224, no. 2019, pp. 425–430, 2023, doi: 10.1016/j.procs.2023.09.059.

[27] A. Qaroush, S. Yassin, A. Al-Nubani, and A. Alqam, “Smart, comfortable wearable system for recognizing Arabic Sign Language in real-time using IMUs and features-based fusion,” Expert Syst. Appl., vol. 184, no. October 2020, p. 115448, 2021, doi: 10.1016/j.eswa.2021.115448.

[28] F. M. Najib, “A multi-lingual sign language recognition system using machine learning,” Multimed. Tools Appl., vol. 84, no. 24, pp. 27987–28011, 2025, doi: 10.1007/s11042-024-20165-3.

[29] T. Aujeszky and M. Eid, “A gesture Recognition architecture for Arabic sign language communication system,” Multimed. Tools Appl., vol. 75, no. 14, pp. 8493–8511, 2016, doi: 10.1007/s11042-015-2767-2.

[30] M. Halabi and Y. Harkouss, “Real-time arabic sign language recognition system using sensory glove and machine learning,” Neural Comput. Appl. 2025 379, vol. 37, no. 9, pp. 6977–6993, 2025, [Online]. Available: https://link.springer.com/article/10.1007/s00521-025-11010-1

[31] R. E. Al Mamlook and A. Aljohani, “Advancing Arabic Sign Language Recognition: A Novel MobileNetv2-Based DL Framework with Superior Accuracy and Cross-Dataset Validation,” Arab. J. Sci. Eng. 2025 5023, vol. 50, no. 23, pp. 20163–20183, 2025, [Online]. Available: https://link.springer.com/article/10.1007/s13369-025-10528-9

[32] A. Boukdir, M. Benaddy, A. Ellahyani, O. El Meslouhi, and M. Kardouchi, “Isolated Video-Based Arabic Sign Language Recognition Using Convolutional and Recursive Neural Networks,” Arab. J. Sci. Eng., vol. 47, no. 2, pp. 2187–2199, 2022, doi: 10.1007/s13369-021-06167-5.

[33] R. Cui, H. Liu, and C. Zhang, “Recurrent convolutional neural networks for continuous sign language recognition by staged optimization,” Proc. - 30th IEEE Conf. Comput. Vis. Pattern Recognition, CVPR 2017, vol. 2017-Janua, pp. 1610–1618, 2017, doi: 10.1109/CVPR.2017.175.

[34] S. Ebling and J. Glauert, “Building a Swiss German Sign Language avatar with JASigning and evaluating it among the Deaf community,” Univers. Access Inf. Soc., vol. 15, no. 4, pp. 577–587, 2016, doi: 10.1007/s10209-015-0408-1.

[35] F. M. Talaat, W. El-Shafai, N. F. Soliman, A. D. Algarni, F. E. Abd El-Samie, and A. I. Siam, “Real-time Arabic avatar for deaf-mute communication enabled by deep learning sign language translation,” Comput. Electr. Eng., vol. 119, no. June, p. 109475, 2024, doi: 10.1016/j.compeleceng.2024.109475.

[36] D. T. Mosa, N. A. Nasef, M. A. Lotfy, A. A. Abohany, R. M. Essa, and A. Salem, “A real-time Arabic avatar for deaf–mute community using attention mechanism,” Neural Comput. Appl., vol. 35, no. 29, pp. 21709–21723, 2023, doi: 10.1007/s00521-023-08858-6.

[37] O. H. Al-Barahamtoshy and H. M. Al-Barhamtoshy, “Arabic Text-to-Sign (ArTTS) Model from Automatic SR System,” Procedia Comput. Sci., vol. 117, pp. 304–311, 2017, doi: 10.1016/j.procs.2017.10.122.

[38] M. Rokaya, D. I. Hemdan, M. A. Alzain, I. Gad, and E. S. Atlam, “Self-supervised learning with a contrastive VideoMoCo framework for Saudi Arabic sign language recognition using 3D convolutional networks,” Sci. Rep., vol. 15, no. 1, pp. 1–18, 2025, doi: 10.1038/s41598-025-23494-x.

[39] A. Boulesnane, L. Bellil, and M. G. Ghiri, “A hybrid CNN-random forest model with landmark angles for real-time Arabic sign language recognition,” Neural Comput. Appl. 2024 374, vol. 37, no. 4, pp. 2641–2662, 2024.

[40] A. A. I. Sidig, H. Luqman, and S. A. Mahmoud, “Transform-based Arabic sign language recognition,” Procedia Comput. Sci., vol. 117, pp. 2–9, 2017, doi: 10.1016/j.procs.2017.10.087.

[41] B. B. Al-Onazi et al., “Arabic Sign Language Gesture Classification Using Deer Hunting Optimization with Machine Learning Model,” Comput. Mater. Contin., vol. 75, no. 2, pp. 3413–3429, 2023, doi: 10.32604/cmc.2023.035303.

[42] E. Aldhahri et al., “Arabic Sign Language Recognition Using Convolutional Neural Network and MobileNet,” Arab. J. Sci. Eng., vol. 48, no. 2, pp. 2147–2154, 2023, doi: 10.1007/s13369-022-07144-2.

[43] M. Aloraini, “ASLR-DPC: Arabic Sign Language Recognition Based on Depthwise and Pointwise Convolutions,” Arab. J. Sci. Eng. 2025 516, vol. 51, no. 6, pp. 7215–7223, 2025, [Online]. Available: https://link.springer.com/article/10.1007/s13369-025-10232-8

[44] P. dos S. Paim and S. S. Prietch, “Problems and Solutions in the Design for Deaf Persons who are Sign Language Users to Adopt Assistive Technology Products,” J. Interact. Syst., vol. 10, no. 2, pp. 70–81, 2019, doi: 10.5753/jis.2019.554.

[45] A. Radford et al., “Learning Transferable Visual Models From Natural Language Supervision,” Proc. Mach. Learn. Res., vol. 139, pp. 8748–8763, 2021.