Performance Evaluation of Intelligent Driving Emotion Recognition Model based on Synthetic Dataset in Real Scenes

Hancheng Li; Yurui Li; Yaowen Qian

doi:10.62051/sae3d060

Authors

Hancheng Li
Yurui Li
Yaowen Qian

DOI:

https://doi.org/10.62051/sae3d060

Keywords:

Intelligent Cockpit; Synthetic Datasets; Emotion Recognition.

Abstract

The paper aims to explore the feasibility of using artificially generated facial expressions with different emotions to enhance the features and increase the data volume for emotion recognition in the context of intelligent cockpit. The paper first introduces the background and significance of the research, which is motivated by the increasing number of private cars in China, the development of intelligent cockpit technology, and the importance of emotion recognition for driving safety. The paper then reviews the existing literature on emotion recognition based on facial recognition, and points out the challenges and limitations of using real datasets, such as ImageNet, which may have high cost, low quality, privacy issues, and inaccurate annotations. The study suggests utilizing synthetic facial expressions that convey a range of emotions, created through advanced deep learning algorithms, as a solution to enhance the precision and reliability of the emotion detection system. The paper further examines the prospective applications and effects of the recommended technique pertaining to the realm of intelligent automotive cockpits and the associated vehicular journey. The paper concludes by summarizing the main contributions and limitations of the research, and suggesting some directions for future work.

Downloads

Download data is not yet available.

References

Nbs. Statistical Communiqué of the People's Republic of China on National Economic and Social Development in 2020 [EB/OL] (2021).

Li H, Wang X, Yu S. Development trend of human-computer interaction in intelligent cockpit[J].Times Automobile, 23 (2022):16-18.

Li W, Cui Y, Ma Y, et al. A Spontaneous Driver Emotion Facial Expression (DEFE) Dataset for Intelligent Vehicles: Emotions Triggered by Video-Audio Clips in Driving Scenarios[J]. IEEE Transactions on Affective Computing, (2021) 14.

Mesken J, Hagenzieker M P, Rothengatter T, et al. Frequency, determinants, and consequences of different drivers’ emotions: An on-the-road study using self-reports, (observed) behaviour, and physiology[J]. Transportation research part F: traffic psychology and behaviour, 10(6) (2007): 458–475.

Organization WH. Global status report on road safety 2015[M]. World Health Organization, (2015).

James L. Road rage and aggressive driving: Steering clear of highway warfare[M]. Prometheus Books, (2000).

Zepf, S.; Hernandez, J.; Schmitt, A.; Minker, W.; Picard, R. Driver emotion recognition for intelligent vehicles: A survey. ACM Comput. Surv. (CSUR) 53 (2020), 1–30.

Li L, Yu H, Yi Z, et al. Quantitative Measure of the Value of Scientific Datasets from Perspective of Altmetrics[J].Information Studies: Theory & Application, 9 (2020):47-52.

Zhang T, Wang C, Yu C, et al.Scientific Research Literature Dataset in the Era of AI: Characteristic Laws and Development Directions[J].Library and Information Knowledge, 40(05) (2023):39-49.

Deng J, Dong W, Socher R, Li J, Li K & Li F, "ImageNet: A large-scale hierarchical image database," 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, FL, USA, (2009) 248-255.

J. Buolamwiniand, T. Gebru. Gendershades: Intersectional accuracy disparities in commercial gender classification[C]. In: S. A. Friedler and C. Wilson, (eds.). Proceedings of Proceedings of the 1st Conference on Fairness, Accountability and Transparency, volume 81 of Proceedings of Machine Learning Research, New York, NY, USA: (2018). 7791.

S. Xie, R. Girshick, P. Dollar, Z. Tu, and K. He. Aggregated residual transformations for deep neural networks[C]. Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), (2017).

Wu H. Research on deep learning method of image processing based on synthetic dataset[D]. Lanzhou University, (2021).

Zhang X, Zhang F & Xu C, "Joint Expression Synthesis and Representation Learning for Facial Expression Recognition," in IEEE Transactions on Circuits and Systems for Video Technology, 32(3) (2022 )1681-1695.

Niinuma, K., Ertugrul, I. O., Cohn, J. F., & Jeni, L. A. Synthetic Expressions are Better Than Real for Learning to Detect Facial Actions. IEEE Winter Conference on Applications of Computer Vision. IEEE Winter Conference on Applications of Computer Vision, (2021), 1247–1256.

Han S, Guo Y, Zhou X, Huang J, Shen L, Luo Y. A Chinese Face Dataset with Dynamic Expressions and Diverse Ages Synthesized by Deep Learning. Sci Data. 10(1) (2023):878.

FER-stable-diffusion-dataset FER 2013.