Comparative Analysis of Deep Neural Networks for Online Facial Expression Recognition and Facial Shape Classification

Sirwan Younis Abdullah; Mayyadah Ramiz Mahmood

doi:10.65542/djei.v2i3.80

Authors

Sirwan Younis Abdullah Information Technology Department, Technical College of Informatics, Akre University for Applied Sciences, Duhok, Iraq
Mayyadah Ramiz Mahmood Computer Science Department, Collage of Science, University of Zakho

DOI:

https://doi.org/10.65542/djei.v2i3.80

Keywords:

Deep learning, facial-expression recognition, face-shape classification, InceptionV3, EfficientNetB0, online facial analysis

Abstract

In the context of affective computing, real-time facial analysis plays an important role in human–computer interaction and personalized applications. This study presents a controlled comparative evaluation of five widely used deep neural network architectures EfficientNetB0, VGG16, MobileNetV2, InceptionV3, and ResNet50 for two classification tasks: recognizing eight facial-expression categories and seven face-shape categories using the mtr_face-v6 dataset. An online recognition pipeline was developed to investigate the trade-off between classification performance and computational efficiency. The experimental results demonstrated that InceptionV3 achieved the highest facial-expression recognition accuracy of 97.4%, benefiting from its inherent multi-scale feature-extraction capability. EfficientNetB0 achieved the highest face-shape classification accuracy of 97.0%, owing to its compound scaling mechanism and ability to learn global facial geometry. Although MobileNetV2 produced lower accuracies of 90.0% for facial-expression recognition and 88.4% for face-shape classification, it provided the lowest inference latency and remained suitable for resource-constrained online applications. Based on these findings, a task-specialized dual-model framework was adopted, integrating InceptionV3 for facial-expression recognition and EfficientNetB0 for face-shape classification within a unified graphical interface. Overall, the results demonstrate that task-specific model selection can achieve high classification performance for both facial-expression recognition and face-shape classification. Future work will investigate a unified multi-task learning architecture to jointly optimize both tasks within a shared framework. This study provides practical guidance for developers and researchers in selecting suitable deep-learning backbones according to the accuracy and computational requirements of online facial-analysis systems.

References

1. Dritsas, E.; Trigka, M.; Troussas, C.; Mylonas, P. Multimodal interaction, interfaces, and communication: A survey. Multimodal Technol. Interact. 2025, 9, 6. DOI: https://doi.org/10.3390/mti9010006

2. Ali, M.; Khan, S.U.R.; Mashkoor, A.; Taskeen, A. A conceptual framework for context-driven self-adaptive intelligent user interface based on Android. Cogn. Technol. Work 2024, 26, 83–106. DOI: https://doi.org/10.1007/s10111-023-00749-z

3. Leong, S.C.; Tang, Y.M.; Lai, C.H.; Lee, C.K.M. Facial expression and body gesture emotion recognition: A systematic review on the use of visual data in affective computing. Comput. Sci. Rev. 2023, 48, 100545. DOI: https://doi.org/10.1016/j.cosrev.2023.100545

4. Jayaswal, R.; Ansari, M.A.; Dixit, M.; Singh, D.K.; Ahmad, S. Advances in facial expression recognition technologies for emotion analysis. Discov. Comput. 2025, 28, 203. DOI: https://doi.org/10.1007/s10791-025-09699-8

5. Wolniak, R.; Stecuła, K. Artificial intelligence in smart cities—Applications, barriers, and future directions: A review. Smart Cities 2024, 7, 1346–1389. DOI: https://doi.org/10.3390/smartcities7030057

6. Xiao, L.; Wu, D.; Ding, M. A smart ad display system. Inf. Syst. Res. 2024, 35, 1873–1889. DOI: https://doi.org/10.1287/isre.2020.0128

7. Krumhuber, E.G.; Skora, L.I.; Hill, H.C.; Lander, K. The role of facial movements in emotion recognition. Nat. Rev. Psychol. 2023, 2, 283–296. DOI: https://doi.org/10.1038/s44159-023-00172-1

8. Valladares, J.; Lopez-Valdes, D.A.; Farrera, A. Reliability of Pöch’s facial shape classification system: A morphometric approach. Forensic Sci. Int. 2025, 369, 112435. DOI: https://doi.org/10.1016/j.forsciint.2025.112435

9. Wang, H.; Song, H.; Li, P. Multi-task network with inter-task consistency learning for face parsing and facial expression recognition at real-time speed. J. Vis. Commun. Image Represent. 2024, 103, 104213. DOI: https://doi.org/10.1016/j.jvcir.2024.104213

10. Gong, Q.; Liu, X.; Ma, Y. Real-time facial expression recognition based on image processing in virtual reality. Int. J. Comput. Intell. Syst. 2025, 18, 8. DOI: https://doi.org/10.1007/s44196-024-00729-9

11. Tian, Y.; Zhu, J.; Yao, H.; Chen, D. Facial expression recognition based on vision transformer with hybrid local attention. Appl. Sci. 2024, 14, 6471. DOI: https://doi.org/10.3390/app14156471

12. Song, D.; Liu, C. A facial expression recognition network using hybrid feature extraction. PLoS ONE 2025, 20, e0312359. DOI: https://doi.org/10.1371/journal.pone.0312359

13. Ibrahem, A.H.; Abdulazeez, A.M. A comprehensive review of facial beauty prediction using multi-task learning and facial attributes. ARO-Sci. J. Koya Univ. 2025, 13, 10–21. DOI: https://doi.org/10.14500/aro.11850

14. Ibrahem, A.H.; Abdelazeez, A.M. KurdFace-1000: A new multi-attribute dataset for facial beauty prediction using multi-task learning. ARO-Sci. J. Koya Univ. 2026, 14, 1–11. DOI: https://doi.org/10.14500/aro.12393

15. Dang, Z.; Li, J.; Liu, L. Cascaded dual vision transformer for accurate facial landmark detection. In Proceedings of the 2025 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV); IEEE: Piscataway, NJ, USA, 2025; pp. 5884–5894. DOI: https://doi.org/10.1109/WACV61041.2025.00574

16. Zhao, J.; Zhu, T.; Sun, L.; Du, B.; Wang, H.; Huang, L. Dual-mask: Progressively sparse multi-task architecture learning. Pattern Recognit. 2025, 158, 110950. DOI: https://doi.org/10.1016/j.patcog.2024.110950

17. Zhao, Y.; et al. H4H: Hybrid convolution-transformer architecture search for NPU-CIM heterogeneous systems for AR/VR applications. In Proceedings of the 30th Asia and South Pacific Design Automation Conference (ASP-DAC), 2025; pp. 1133–1141. DOI: https://doi.org/10.1145/3658617.3697627

18. Kopalidis, T.; Solachidis, V.; Vretos, N.; Daras, P. Advances in facial expression recognition: A survey of methods, benchmarks, models, and datasets. Information 2024, 15, 135. DOI: https://doi.org/10.3390/info15030135

19. Vandenhende, S.; Georgoulis, S.; Van Gansbeke, W.; Proesmans, M.; Dai, D.; Van Gool, L. Multi-task learning for dense prediction tasks: A survey. IEEE Trans. Pattern Anal. Mach. Intell. 2022, 44, 3614–3633. DOI: https://doi.org/10.1109/TPAMI.2021.3054719

20. Liu, L.; Xu, Z. Optimizing lightweight neural networks for efficient mobile edge computing. Sci. Rep. 2025, 15, 22056. DOI: https://doi.org/10.1038/s41598-025-04652-7

21. Grd, P.; Tomičić, I.; Barčić, E. Transfer learning with EfficientNetV2S for automatic face shape classification. J. Univers. Comput. Sci. 2024, 30, No. 2. DOI: https://doi.org/10.3897/jucs.104490

22. Hu, G.; et al. Deep multi-task learning to recognise subtle facial expressions of mental states. In Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany, 8–14 September 2018; pp. 103–119. DOI: https://doi.org/10.1007/978-3-030-01258-8_7

23. Zhang, K.; Zhang, Z.; Li, Z.; Qiao, Y. Joint face detection and alignment using multitask cascaded convolutional networks. IEEE Signal Process. Lett. 2016, 23, 1499–1503. DOI: https://doi.org/10.1109/LSP.2016.2603342

24. MTR Face Dataset V6. Kaggle. Available online: https://www.kaggle.com/datasets/mtrakretech/mtr-face-dataset-v6 (accessed on 8 August 2026).

25. Simonyan, K.; Zisserman, A. Very deep convolutional networks for large-scale image recognition. arXiv 2014, arXiv:1409.1556.

26. Shafiq, M.; Gu, Z. Deep residual learning for image recognition: A survey. Appl. Sci. 2022, 12, 8972. DOI: https://doi.org/10.3390/app12188972

27. Agung, E.S.; Rifai, A.P.; Wijayanto, T. Image-based facial emotion recognition using convolutional neural network on Emognition dataset. Sci. Rep. 2024, 14, 14429. https://doi.org/10.1038/s41598-024-65276-x. DOI: https://doi.org/10.1038/s41598-024-65276-x