EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Adaptive 3D Human Pose Estimation Based on Spatial–Temporal Complexity Awareness

Wensi Zhang, Ziyan Yang, Chengfeng Hu, Jing Sun, Jie Li

Electronics · 13.05.2026 · 15 (10) · 2076

DOI: 10.3390/electronics15102076 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 13.05.2026. Verfasst von Wensi Zhang und 4 weiteren Autor:innen. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Provincial-level College Students’ Innovation and Entrepreneurship Training Program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Existing 3D human pose estimation methods use fixed computation strategies processing diverse action sequences, leading to computational redundancy for simple actions, insufficient high-frequency information capture for complex actions, and low long-sequence processing efficiency. To address these issues, this paper proposes a Spatial–Temporal Complexity-Aware Adaptive Computation Framework (CAAPoseFormer). First, a spatial–temporal coupled complexity quantification module is built to integrate spatial dispersion and temporal motion variance for graded action complexity quantification. On this basis, a time–frequency dual-domain adaptive pruning strategy is proposed to dynamically allocate temporal window length and frequency-domain DCT coefficients on demand. Furthermore, a mask-guided sparse interaction encoding mechanism is designed to enable efficient parallel computation of variable-length features by shielding invalid padding regions. Experiments on the Human3.6M dataset show that, versus the baseline PoseFormerV2, the proposed method cuts parameters by 85.3% and computational cost by 64.8% while retaining comparable accuracy (MPJPE 44.2 mm), boosting unit computational efficiency 2.8×. Moreover, compared with state-of-the-art (SOTA) methods like MHFormer and MotionBERT, our method reduces computational costs (MACs) by 97.4% and nearly three orders of magnitude, respectively. This framework effectively breaks the inference bottleneck of high-precision models on low-power hardware, suiting latency-sensitive real-time applications well.

Förderung

  • Provincial-level College Students’ Innovation and Entrepreneurship Training Program

Zitieren

APA

Zhang, W., Yang, Z., Hu, C., Sun, J., & Li, J. (2026). Adaptive 3D Human Pose Estimation Based on Spatial–Temporal Complexity Awareness. Electronics, 15(10), 2076. https://doi.org/10.3390/electronics15102076

Vancouver

Zhang W, Yang Z, Hu C, Sun J, Li J. Adaptive 3D Human Pose Estimation Based on Spatial–Temporal Complexity Awareness. Electronics. 2026;15(10):2076. doi:10.3390/electronics15102076

BibTeX
@article{Zhang2026,
  author = {Zhang, Wensi and Yang, Ziyan and Hu, Chengfeng and Sun, Jing and Li, Jie},
  title = {Adaptive 3D Human Pose Estimation Based on Spatial–Temporal Complexity Awareness},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {10},
  pages = {2076},
  doi = {10.3390/electronics15102076}
}
Kostenlos starten →