EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

ADCCR: Adaptive Description Composition and Coarse-to-Fine Refinement for Language-Guided Human Keypoint Localization

Kuokuo Zhang, Nan Xu, Xi Li, Hai Liu, Fei Wang, Zhenhua Xiao

Biomimetics · 31.10.2026 · 11 (11) · 779

DOI: 10.3390/biomimetics11110779 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biomimetics, online erschienen am 31.10.2026. Verfasst von Kuokuo Zhang und 5 weiteren Autor:innen. Die Arbeit nennt 58 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Jiangxi Provincial Natural Science Foundation, Science and Technology Research Project of Jiangxi Provincial Department of Education.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Biological vision integrates cues according to their reliability and context and uses coarse contextual information to form initial hypotheses before directing attention to local evidence. Inspired by these principles, we propose ADCCR (Adaptive Description Composition and Coarse-to-Fine Refinement), a language-guided framework for flexible human keypoint localization. A multi-granularity bank describes each keypoint by its name, anatomical definition, inter-joint relation, and visual appearance. A reliability-aware composer predicts sample-specific gates from the person image, target identity, and candidate cues, replacing predefined sampling probabilities. Counterfactual cue-consistency supervision suppresses inconsistent cues, while a sparse cue-budget regularizer reduces redundancy. The selected description guides a multimodal large language model to estimate a coarse coordinate from the whole-person image. A category-aware crop of the resized person image then focuses text-conditioned heatmap refinement on the candidate region. ADCCR thus provides a functional abstraction of reliability-aware cue integration and selective global-to-local inspection. Trained on COCO, ADCCR achieves 80.1 AP on the validation set, outperforming PoseLLM by 2.5 points. Results on MPII and Human-Art further demonstrate strong cross-dataset generalization.

Förderung

  • National Natural Science Foundation of China
  • Jiangxi Provincial Natural Science Foundation
  • Science and Technology Research Project of Jiangxi Provincial Department of Education

Zitieren

APA

Zhang, K., Xu, N., Li, X., Liu, H., Wang, F., & Xiao, Z. (2026). ADCCR: Adaptive Description Composition and Coarse-to-Fine Refinement for Language-Guided Human Keypoint Localization. Biomimetics, 11(11), 779. https://doi.org/10.3390/biomimetics11110779

Vancouver

Zhang K, Xu N, Li X, Liu H, Wang F, Xiao Z. ADCCR: Adaptive Description Composition and Coarse-to-Fine Refinement for Language-Guided Human Keypoint Localization. Biomimetics. 2026;11(11):779. doi:10.3390/biomimetics11110779

BibTeX
@article{Zhang2026,
  author = {Zhang, Kuokuo and Xu, Nan and Li, Xi and Liu, Hai and Wang, Fei and Xiao, Zhenhua},
  title = {ADCCR: Adaptive Description Composition and Coarse-to-Fine Refinement for Language-Guided Human Keypoint Localization},
  journal = {Biomimetics},
  year = {2026},
  volume = {11},
  number = {11},
  pages = {779},
  doi = {10.3390/biomimetics11110779}
}
Kostenlos starten →