Bridging the Gap in Facial Expression Recognition: Real-Time Use and Human-in-the-Loop Data Collection

• Paulo Ricardo UEA
• Tiago de Melo UEA
• Elloá B. Guedes UEA

DOI: https://doi.org/10.5753/eniac.2026.29853

Resumo

A avaliação tradicional de soluções de reconhecimento de expressões faciais (REF) não reflete os desafios de sua utilização em situações reais. Para mitigar essa limitação, este artigo propõe um sistema web de REF em tempo real (YOLOv8n, WebRTC, FastAPI) integrado a um mecanismo human-in-the-loop que converte os erros de classificação em amostras rotuladas no formato YOLO, compilando exemplos difíceis para futuros re-treinos. A avaliação empírica confirmou a lacuna entre avaliação de uma solução REF em benchmark em contraste com um cenário de uso prático, em que a vazão de 56 FPS reduziu para 25 FPS (GPU) e 7 FPS (CPU) no fluxo completo, revelando também sensibilidade à pose e à qualidade da captura. O sistema proposto viabiliza a coleta contínua de dados e apoia o aprimoramento futuro do modelo.

Referências

Alshammari, A. and Alshammari, M. E. (2024). Emotional facial expression detection using YOLOv8. Engineering, Technology & Applied Science Research, 14(5):16619–16623.

Jegham, N., Koh, C. Y., Abdelatti, M. F., and Hendawi, A. (2025). Yolo evolution: A comprehensive benchmark and architectural review of yolov12, yolo11, and their previous versions. Preprint.

Jocher, G., Chaurasia, A., and Qiu, J. (2023). Ultralytics yolov8.

Kumari, J., Rajesh, R., and Pooja, K. M. (2015). Facial expression recognition: A survey. Procedia Computer Science, 58:486–491.

Li, W., Wang, J., Zhang, L., Zeng, X., Chen, G., and Long, H. (2024). Research on facial expression recognition based on improved yolov8. In Proc. SPIE 13230, Third International Conference on Machine Vision, Automatic Identification, and Detection (MVAID 2024).

Liao, J., Lin, Y., Ma, T., He, S., Liu, X., and He, G. (2023). Facial expression recognition methods in the wild based on fusion feature of attention mechanism and lbp. PMC, 2023:1–21.

Liu, P., Qi, B., and Banerjee, S. (2018). Edgeeye: An edge service framework for real-time intelligent video analytics. In Proceedings of the 1st International Workshop on Edge Systems, Analytics and Networking, EdgeSys ’18, pages 1–6, New York, NY, USA. Association for Computing Machinery.

Lu, H., Xu, J., and Zheng, H. (2025). Explore the facial expression recognition in different complex environment. Applied and Computational Engineering, 174:77–85.

Mahmoud, H. and Abozariba, R. (2025). A systematic review on webrtc for potential applications and challenges beyond audio video streaming. Multimedia Tools and Applications, 84:2909–2946.

Mollahosseini, A., Hasani, B., and Mahoor, M. H. (2017). Affectnet: A database for facial expression, valence, and arousal computing in the wild. IEEE Transactions on Affective Computing, 10(1):18–31.

Redmon, J., Divvala, S., Girshick, R., and Farhadi, A. (2016). You only look once: Unified, real-time object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 779–788.

Ricardo, P. and de Melo, T. (2026). Reconhecimento de expressoes faciais em tempo real com yolo: Estudo comparativo em cenários nao controlados. In Seminário Integrado de Software e Hardware (SEMISH), pages 998–1003. SBC.

Sareen, V. and Seeja, K. R. (2025). Video-based facial emotion recognition using YOLO and Vision Transformer. In EPJ Web of Conferences, volume 328, page 01040, Les Ulis, France. EDP Sciences. Published online: 18 June 2025.

Sassu, A., Saenz-Cogollo, J. F., and Agelli, M. (2021). Deep-framework: A distributed, scalable, and edge-oriented framework for real-time analysis of video streams. Sensors, 21(12):4045.

Shao, J. and Qian, Y. (2019). Three convolutional neural network models for facial expression recognition in the wild. Neurocomputing, 355:82–92.

Taware, S. and Thakare, A. (2023). Critical analysis on multimodal emotion recognition in meeting the requirements for next generation human computer interactions. International Journal on Recent and Innovation Trends in Computing and Communication, 11(9s):523–531.

Wu, T. (2024). Reducing the cost of test data labelling for deep-learning systems: An empirical study. Master’s thesis, School of Electrical Engineering and Computer Science, University of Ottawa, Ottawa, Ontario, Canada. Disponível em DOI: 10.20381/ruor-30788. Acesso em: 15 de maio de 2026.

Zhong, H., Han, T., Xia, W., Tian, Y., and Wu, L. (2023). Research on real-time teachers’ facial expression recognition based on yolov5 and attention mechanisms. EURASIP Journal on Advances in Signal Processing, 2023(55):1–17.