Research on Pupil Center Localization Detection Algorithm with Improved YOLOv8
Source abstract
Abstract
Addressing issues such as low localization accuracy, poor robustness, and long average localization time in pupil center localization algorithms, an improved YOLOv8 network-based pupil center localization algorithm is proposed. This algorithm incorporates a dual attention mechanism into the YOLOv8n backbone network, which simultaneously attends to global contextual information of input data while reducing dependence on specific regions. This improves the problem of difficult pupil localization detection due to occlusions such as eyelashes and eyelids, enhancing the model’s robustness. Additionally, atrous convolutions are introduced in the encoding section, which reduce the network model while improving the model’s detection speed. The use of the Focaler-IoU loss function, by focusing on different regression samples, can improve the performance of detectors in various detection tasks. The performance of the improved Yolov8n algorithm was 0.99971, 1, 0.99611, and 0.96495 in precision, recall, MAP50, and mAP50-95, respectively. Moreover, the improved YOLOv8n algorithm reduced the model parameters by 7.18% and the computational complexity by 10.06%, while enhancing the environmental anti-interference ability and robustness, and shortening the localization time, improving real-time detection.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Gaze Tracking and Assistive Technology · Ocular Surface and Contact Lens · Face recognition and analysis
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Optoelectronic Engineering, Changchun University of Science and Technology, Changchun 130022, China
Zhongshan Institute of Changchun University of Science and Technology, Zhongshan 528437, China
Funding
Jilin Provincial Science and Technology Development Program Key R&D Project · Jilin Province Key Science and Technology Research and Development projects · Jilin Provincial Department of Education’s 13th Five-Year Plan Science and Technology Project
References (28)
Show source-provided references
1. Tracking gaze and visual focus of attention of people involved in social interaction10.1109/TPAMI.2017.2782819 ↗
2. Current Trends in Human Pupil Localization: A Review10.1109/ACCESS.2023.3325293 ↗
3. An effective iris segmentation scheme for noisy images
4. Poulopoulos, N., and Psarakis, E. (2022, January 6–8). DeepPupil net: Deep residual network for precise pupil center localization. Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2022), Setúbal, Portugal.10.5220/0010777900003124 ↗
5. Accurate Eye Centre Localisation by Means of Gradient
6. Nugroho, R.H., Nasrun, M., and Setianingsih, C. (2017, January 26–28). Lie Detector with Pupil Dilation and Eye Blinks Using Hough Transform and Frame Difference Method with Fuzzy Logic. Proceedings of the 2017 International Conference on Control, Electronics, Renew−able Energyand Communication (ICCREC), Yogyakarta, Indonesia.10.1109/ICCEREC.2017.8226697 ↗
7. High−precision Pupil Center Positioning Method Based on Near−eye Infrared Image
8. Pupil location algorithm applied to infrared ophthalmic disease detection10.37188/CO.2020-0170 ↗
9. Gaze direction estimation by component separation for recognition of Eye Accessing Cues10.1007/s00138-014-0656-8 ↗
10. Research and implementation of pupil recognition based on Hough transform10.3788/YJYXS20163106.0621 ↗
11. Fast radial symmetry for detecting points of interest10.1109/TPAMI.2003.1217601 ↗
12. Research on pupil localization algorithm based on gradient direction constraint
13. Pupil Nte: Convolutional neural networks for robust pupil detection
14. CNN−based pupil center detection for wearable gaze estimation system
15. Pupil center location based on star ray method
16. Fuhl, W., Eivazi, S., Hosp, B., Eivazi, A., Rosenstiel, W., and Kasneci, E. (2018, January 14–17). BORE: Boosted-oriented edge optimization for robust, real time remote pupil center detection. Proceedings of the 2018 ACM Symposium on Eye Tracking Research & Applications, Warsaw, Poland.10.1145/3204493.3204558 ↗
17. RemoteEye: An open-source high-speed remote eye tracker: Implementation insights of a pupil- and glint-detection algorithm for high-speed remote eye tracking10.3758/s13428-019-01305-2 ↗
18. Donuk, K., and Hanbay, D. (2022, January 22). Pupil center localization based on mini U-net. Proceedings of the International Artificial Intelligence and Data Processing Symposium, (IDAP), Sofia, Bulgaria.
19. Chen, G., Dong, Z., Wang, J., and Xia, L. (2023). ‘Pupil localization algorithm based on improved U-Net network. Electronics, 12.10.3390/electronics12122591 ↗
20. Research on pupil location algorithm of non−contact tonometer
21. Pupil location algorithm based on Attention Gate and dilated convolution
22. Review of Typical Object Detection Algorithms for Deep Learning
23. Li, X. (2024). Research on Pupil Center Location and Tracking Based on Near−Eye Infrared Video Images. [Master’s Thesis, North University of China].
24. Improved road defect detection algorithm based on YOLOv8
25. Zhang, L., Sun, Z., Tao, H., Hao, S., Yan, Q., and Li, X. (2024). Research on real−time monitoring method of mine personnel protective equipment with improved YOLOv8. Coal Sci. Technol., 1–12. Available online: https://link.cnki.net/urlid/11.2402.td.20240527.1700.004.
26. Safety Helmet Detection Based on Improved YOLOv810.1109/ACCESS.2024.3368161 ↗
27. Improved YOLOv8 Object Detection Algorithm for Traffic Sign Target
28. Hu, M., Wang, R., Zhang, W., and Zhang, Q. (2024). Multi−scale Referring Image Segmentation Based on Dual Attention. J. Comput.-Aided Des. Comput. Graph., 1–10. Available online: https://link.cnki.net/urlid/11.2925.tp.20240531.1456.002.