YOLOv8-ACU: improved YOLOv8-pose for facial acupoint detection
IntroductionAcupoint localization is integral to Traditional Chinese Medicine (TCM) acupuncture diagnosis and treatment. Employing intelligent detection models for recognizing facial acupoints can substantially enhance localization accuracy.MethodsThis study introduces an advancement in the YOLOv8-p...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-02-01
|
| Series: | Frontiers in Neurorobotics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnbot.2024.1355857/full |
| _version_ | 1797337241380454400 |
|---|---|
| author | Zijian Yuan Pengwei Shao Jinran Li Yinuo Wang Zixuan Zhu Weijie Qiu Buqun Chen Yan Tang Aiqing Han |
| author_facet | Zijian Yuan Pengwei Shao Jinran Li Yinuo Wang Zixuan Zhu Weijie Qiu Buqun Chen Yan Tang Aiqing Han |
| author_sort | Zijian Yuan |
| collection | DOAJ |
| description | IntroductionAcupoint localization is integral to Traditional Chinese Medicine (TCM) acupuncture diagnosis and treatment. Employing intelligent detection models for recognizing facial acupoints can substantially enhance localization accuracy.MethodsThis study introduces an advancement in the YOLOv8-pose keypoint detection algorithm, tailored for facial acupoints, and named YOLOv8-ACU. This 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.ResultsThe YOLOv8-ACU model achieves impressive accuracy, with an mAP@0.5 of 97.5% and an mAP@0.5–0.95 of 76.9% on our self-constructed datasets. It also marks a reduction in model parameters by 0.44M, model size by 0.82 MB, and GFLOPs by 9.3%.DiscussionWith its enhanced recognition accuracy and efficiency, along with good generalization ability, YOLOv8-ACU provides significant reference value for facial acupoint localization and detection. This is particularly beneficial for Chinese medicine practitioners engaged in facial acupoint research and intelligent detection. |
| first_indexed | 2024-03-08T09:07:40Z |
| format | Article |
| id | doaj.art-b2edd40ae69947cfa5a42c638d1f2694 |
| institution | Directory Open Access Journal |
| issn | 1662-5218 |
| language | English |
| last_indexed | 2024-03-08T09:07:40Z |
| publishDate | 2024-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neurorobotics |
| spelling | doaj.art-b2edd40ae69947cfa5a42c638d1f26942024-02-01T04:25:19ZengFrontiers Media S.A.Frontiers in Neurorobotics1662-52182024-02-011810.3389/fnbot.2024.13558571355857YOLOv8-ACU: improved YOLOv8-pose for facial acupoint detectionZijian Yuan0Pengwei Shao1Jinran Li2Yinuo Wang3Zixuan Zhu4Weijie Qiu5Buqun Chen6Yan Tang7Aiqing Han8College of Management, Beijing University of Chinese Medicine, Beijing, ChinaCollege of Management, Beijing University of Chinese Medicine, Beijing, ChinaBeijing No.80 High School, International Department, Beijing, ChinaCollege of Management, Beijing University of Chinese Medicine, Beijing, ChinaCollege of Acupuncture and Massage, Beijing University of Chinese Medicine, Beijing, ChinaCollege of Acupuncture and Massage, Beijing University of Chinese Medicine, Beijing, ChinaCollege of Acupuncture and Massage, Beijing University of Chinese Medicine, Beijing, ChinaCollege of Management, Beijing University of Chinese Medicine, Beijing, ChinaCollege of Management, Beijing University of Chinese Medicine, Beijing, ChinaIntroductionAcupoint localization is integral to Traditional Chinese Medicine (TCM) acupuncture diagnosis and treatment. Employing intelligent detection models for recognizing facial acupoints can substantially enhance localization accuracy.MethodsThis study introduces an advancement in the YOLOv8-pose keypoint detection algorithm, tailored for facial acupoints, and named YOLOv8-ACU. This 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.ResultsThe YOLOv8-ACU model achieves impressive accuracy, with an mAP@0.5 of 97.5% and an mAP@0.5–0.95 of 76.9% on our self-constructed datasets. It also marks a reduction in model parameters by 0.44M, model size by 0.82 MB, and GFLOPs by 9.3%.DiscussionWith its enhanced recognition accuracy and efficiency, along with good generalization ability, YOLOv8-ACU provides significant reference value for facial acupoint localization and detection. This is particularly beneficial for Chinese medicine practitioners engaged in facial acupoint research and intelligent detection.https://www.frontiersin.org/articles/10.3389/fnbot.2024.1355857/fullChinese medicine acupointsYOLOv8-posekeypoint detectionECA-netslim-neckGIoU |
| spellingShingle | Zijian Yuan Pengwei Shao Jinran Li Yinuo Wang Zixuan Zhu Weijie Qiu Buqun Chen Yan Tang Aiqing Han YOLOv8-ACU: improved YOLOv8-pose for facial acupoint detection Frontiers in Neurorobotics Chinese medicine acupoints YOLOv8-pose keypoint detection ECA-net slim-neck GIoU |
| title | YOLOv8-ACU: improved YOLOv8-pose for facial acupoint detection |
| title_full | YOLOv8-ACU: improved YOLOv8-pose for facial acupoint detection |
| title_fullStr | YOLOv8-ACU: improved YOLOv8-pose for facial acupoint detection |
| title_full_unstemmed | YOLOv8-ACU: improved YOLOv8-pose for facial acupoint detection |
| title_short | YOLOv8-ACU: improved YOLOv8-pose for facial acupoint detection |
| title_sort | yolov8 acu improved yolov8 pose for facial acupoint detection |
| topic | Chinese medicine acupoints YOLOv8-pose keypoint detection ECA-net slim-neck GIoU |
| url | https://www.frontiersin.org/articles/10.3389/fnbot.2024.1355857/full |
| work_keys_str_mv | AT zijianyuan yolov8acuimprovedyolov8poseforfacialacupointdetection AT pengweishao yolov8acuimprovedyolov8poseforfacialacupointdetection AT jinranli yolov8acuimprovedyolov8poseforfacialacupointdetection AT yinuowang yolov8acuimprovedyolov8poseforfacialacupointdetection AT zixuanzhu yolov8acuimprovedyolov8poseforfacialacupointdetection AT weijieqiu yolov8acuimprovedyolov8poseforfacialacupointdetection AT buqunchen yolov8acuimprovedyolov8poseforfacialacupointdetection AT yantang yolov8acuimprovedyolov8poseforfacialacupointdetection AT aiqinghan yolov8acuimprovedyolov8poseforfacialacupointdetection |