EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Occlusion-Robust Cattle Pose Estimation for Precision Livestock Monitoring Using Hierarchical Locality Refinement

Yingchao Wang iD, Na Li iD, Dan He, Shan Sun, Xinjian Chu, Zixiang Qin, Feng Xue iD, Jingjun Yi iD, Hao Wu, Han-Su Zhang, Fan Zhao iD

Animals · 18.08.2026 · 16 (16) · 2575

DOI: 10.3390/ani16162575 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Animals, online erschienen am 18.08.2026. Verfasst von Yingchao Wang und 10 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science and Technology, Xinjiang College of Science & Technology, Korla 841000, China. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 78 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Xinjiang Uygur Autonomous Region "Tianchi Talents" Introduction Program Innovation Leading Talent Project, Industry-Education Integration and New Business Development Research Center Project, Xinjiang Uygur Autonomous Region “Tianchi Talents” Introduction Program Innovation Leading Talent Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate cattle pose estimation is important for precision livestock farming because body landmarks can provide quantitative visual inputs for potential health monitoring, behavior analysis, lameness assessment, and welfare evaluation. This study focuses on keypoint-localization performance and provides a foundation for future task-specific studies of these downstream outcomes. However, real farm images commonly contain inter-cattle occlusion, cluttered backgrounds, small anatomical landmarks, and visually similar animals, which limit the reliability of existing one-stage pose estimators. This study proposes LEC-Pose, a hierarchical locality refinement framework for occlusion-robust cattle pose estimation. Built on YOLOv8-Pose, LEC-Pose first predicts cattle boxes and auxiliary coarse keypoints, then extracts instance-level ROI features to recover local anatomical evidence. A lightweight refinement network directly predicts the final keypoints from heatmaps and offsets on enhanced ROI features. The detected boxes guide the ROI pathway, while the initial keypoints remain auxiliary outputs. During training, an instance-level contrastive loss regularizes a compact global ROI descriptor. During inference, pair construction and contrastive-loss computation are removed, while descriptor fusion remains in the prediction path. Across three runs on CattleEyeView, LEC-Pose reaches 39.47 ± 0.33 mAP@50:95, compared with 33.91 ± 0.28 for YOLOv8-Pose, while running at 111.6 FPS versus 138.2 FPS for YOLOv8-Pose, corresponding to a 19.2% throughput reduction on the evaluated RTX 4090. On NWAFU-Cattle, it obtains 76.08 ± 0.36 and 80.03 ± 0.38 mAP@50:95 under the 50%/50% and 80%/20% protocols, respectively; the former is comparable to FSMC-Pose, whereas the latter is the highest mean among the three repeated principal models. An end-to-end zero-shot evaluation on three external cattle datasets, second-dataset occlusion analysis, detailed module combinations, and descriptor diagnostics further assess the framework within their stated protocols. These results suggest that LEC-Pose can serve as a pose-estimation component for future automated cattle monitoring, and future task-specific studies can connect these keypoints to health, behavior, and welfare outcomes.

Schlagwörter

Stichwörter der Autor:innen

ROI refinementcattle pose estimationhierarchical locality refinementinstance-level contrastive regularizationkeypoint localizationprecision livestock farming

Autor:innen und Institutionen

  • Department of Computer Science and Technology, Xinjiang College of Science & Technology, Korla 841000, China.
  • College of Oceanography and Space Informatics, China University of Petroleum (East China), Qingdao 266580, China.
  • Graduate School of Information, Production and Systems, Waseda University, Kitakyushu 808-0135, Japan.
  • State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430079, China.
  • Graduate School of Information Science and Technology, The University of Tokyo, Tokyo 113-8656, Japan.
  • School of Artificial Intelligence and Information Technology, Nanjing University of Chinese Medicine, Nanjing 210023, China.

ORCID-Profile: Yingchao Wang, Na Li, Feng Xue, Jingjun Yi, Fan Zhao

Förderung

  • Xinjiang Uygur Autonomous Region "Tianchi Talents" Introduction Program Innovation Leading Talent Project
  • Industry-Education Integration and New Business Development Research Center Project
  • Xinjiang Uygur Autonomous Region “Tianchi Talents” Introduction Program Innovation Leading Talent Project
  • Xinjiang Institute of Science and Technology University-level Research Fund Project

Zitieren

APA

Wang, Y., Li, N., He, D., Sun, S., Chu, X., Qin, Z., Xue, F., Yi, J., Wu, H., Zhang, H. S., & Zhao, F. (2026). Occlusion-Robust Cattle Pose Estimation for Precision Livestock Monitoring Using Hierarchical Locality Refinement. Animals, 16(16), 2575. https://doi.org/10.3390/ani16162575

Vancouver

Wang Y, Li N, He D, Sun S, Chu X, Qin Z, et al. Occlusion-Robust Cattle Pose Estimation for Precision Livestock Monitoring Using Hierarchical Locality Refinement. Animals. 2026;16(16):2575. doi:10.3390/ani16162575

BibTeX
@article{Wang2026,
  author = {Wang, Yingchao and Li, Na and He, Dan and Sun, Shan and Chu, Xinjian and Qin, Zixiang and Xue, Feng and Yi, Jingjun and Wu, Hao and Zhang, Han-Su and Zhao, Fan},
  title = {Occlusion-Robust Cattle Pose Estimation for Precision Livestock Monitoring Using Hierarchical Locality Refinement},
  journal = {Animals},
  year = {2026},
  volume = {16},
  number = {16},
  pages = {2575},
  doi = {10.3390/ani16162575}
}
Kostenlos starten →