YOLOv8-ACU: improved YOLOv8-pose for facial acupoint detection

YOLOv8-ACU: improved YOLOv8-pose for facial acupoint detection

01 February 2024 | Zijian Yuan1†, Pengwei Shao1†, Jinran Li2, Yinuo Wang1, Zixuan Zhu3, Weijie Qiu3, Buqun Chen3, Yan Tang1* and Aiqing Han1*
This study introduces YOLOv8-ACU, an advanced version of the YOLOv8-pose keypoint detection algorithm tailored for facial acupoint detection. The model enhances acupoint feature extraction by integrating ECA attention, replaces the original neck module with a lighter Slim-neck module, and improves the loss function for GIoU. The results show that YOLOv8-ACU achieves impressive accuracy with an mAP@0.5 of 97.5% and an mAP@0.5–0.95 of 76.9% on self-constructed datasets, while reducing the model parameters by 0.44M, model size by 0.82 MB, and GFLOPs by 9.3%. The study discusses the benefits of YOLOv8-ACU in terms of recognition accuracy, efficiency, and generalization ability, making it a significant reference for facial acupoint localization and detection in Chinese medicine.This study introduces YOLOv8-ACU, an advanced version of the YOLOv8-pose keypoint detection algorithm tailored for facial acupoint detection. The model enhances acupoint feature extraction by integrating ECA attention, replaces the original neck module with a lighter Slim-neck module, and improves the loss function for GIoU. The results show that YOLOv8-ACU achieves impressive accuracy with an mAP@0.5 of 97.5% and an mAP@0.5–0.95 of 76.9% on self-constructed datasets, while reducing the model parameters by 0.44M, model size by 0.82 MB, and GFLOPs by 9.3%. The study discusses the benefits of YOLOv8-ACU in terms of recognition accuracy, efficiency, and generalization ability, making it a significant reference for facial acupoint localization and detection in Chinese medicine.
Reach us at info@study.space