EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Geometry-Aware Enhanced 6DRepNet for Single-RGB Head Pose Estimation

Hua Yang iD, Yuanyuan Li, Mingzhi Mu iD, Ming Zhao iD

Sensors · 21.05.2026 · 26 (10) · 3266

DOI: 10.3390/s26103266 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 21.05.2026. Verfasst von Hua Yang und 3 weiteren Autor:innen. Beteiligte Einrichtung: College of Mathematics and Computer Science, Wuhan Polytechnic University, Wuhan 430040, China. Die Arbeit nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Hubei Provincial Natural Science Foundation, Ministry of Education Industry-University Cooperation Education Project, 2025 ESI.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Head pose estimation is a fundamental task in facial analysis and behavior understanding. To address the limitations of 6DRepNet in single-RGB scenarios, particularly in high-level spatial discriminative region modeling, global feature-to-6D rotation representation mapping, and the optimization of large-pose challenging samples, this paper proposes a geometry-aware enhanced framework for head pose estimation. While preserving the 6D continuous rotation representation and its SO(3)-based geometric supervision mechanism, the proposed method improves the baseline model through the joint design of a Spatial Recalibration Module, a Residual Pose Mapping Head, and a Pose-Aware Weighted Geodesic Loss. Experiments are conducted using 300W-LP for training and AFLW2000 and BIWI for evaluation. The results show that the proposed method consistently outperforms the baseline 6DRepNet on both datasets, reducing the overall MAE from 3.97 to 3.72 on AFLW2000 and from 3.54 to 3.26 on BIWI. Ablation studies further verify the effectiveness and complementarity of the proposed components. These results demonstrate that the proposed method can effectively improve the accuracy and robustness of single-RGB head pose estimation without introducing additional modalities.

Schlagwörter

Stichwörter der Autor:innen

6D rotation representation6DRepNetgeodesic lossgeometric consistencyhead pose estimationsingle-RGB

Autor:innen und Institutionen

  • College of Mathematics and Computer Science, Wuhan Polytechnic University, Wuhan 430040, China.

ORCID-Profile: Hua Yang, Mingzhi Mu, Ming Zhao

Förderung

  • Hubei Provincial Natural Science Foundation
  • Ministry of Education Industry-University Cooperation Education Project
  • 2025 ESI
  • Research and Application of Multimodal Algorithms
  • Research and Development and Application of Large-Scale Image Retrieval Based on Cloud Platform and Distributed Computing
  • University Graduate Teaching and Research Project, Research on Interdisciplinary Collaborative Innovation and Joint Training Mechanism in Artificial Intelligen…
  • Graduate Student Workstation Project of Wuhan Polytechnic University

Zitieren

APA

Yang, H., Li, Y., Mu, M., & Zhao, M. (2026). Geometry-Aware Enhanced 6DRepNet for Single-RGB Head Pose Estimation. Sensors, 26(10), 3266. https://doi.org/10.3390/s26103266

Vancouver

Yang H, Li Y, Mu M, Zhao M. Geometry-Aware Enhanced 6DRepNet for Single-RGB Head Pose Estimation. Sensors. 2026;26(10):3266. doi:10.3390/s26103266

BibTeX
@article{Yang2026,
  author = {Yang, Hua and Li, Yuanyuan and Mu, Mingzhi and Zhao, Ming},
  title = {Geometry-Aware Enhanced 6DRepNet for Single-RGB Head Pose Estimation},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {10},
  pages = {3266},
  doi = {10.3390/s26103266}
}
Kostenlos starten →