Home > Research > Publications & Outputs > Skeleton-prompt

Electronic data

  • SKELETON_PROMPT

    Accepted author manuscript, 2.65 MB, PDF document

    Available under license: CC BY: Creative Commons Attribution 4.0 International License

Links

Text available via DOI:

View graph of relations

Skeleton-prompt: A cross-dataset transfer learning approach for skeleton action recognition

Research output: Contribution to Journal/MagazineJournal articlepeer-review

E-pub ahead of print

Standard

Skeleton-prompt: A cross-dataset transfer learning approach for skeleton action recognition. / Lu, M.; Lu, X.; Liu, J.
In: Pattern Recognition, Vol. 169, 111885, 31.01.2026.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Vancouver

Lu M, Lu X, Liu J. Skeleton-prompt: A cross-dataset transfer learning approach for skeleton action recognition. Pattern Recognition. 2026 Jan 31;169:111885. Epub 2025 Jun 9. doi: 10.1016/j.patcog.2025.111885

Author

Bibtex

@article{081fb852b5c244da95f8702069751d17,
title = "Skeleton-prompt: A cross-dataset transfer learning approach for skeleton action recognition",
abstract = "This paper presents Skeleton-Prompt, a novel tuning method designed to tackle cross-dataset transfer issues in skeleton action recognition models. Given the scarcity of large-scale 3D skeleton datasets and the variability in keypoint structures across datasets, existing methods often rely on training models from scratch, necessitating extensive labeled data and exhibiting high sensitivity to occlusion. Our approach aims to fine-tune pre-trained models to adapt to limited real-world skeleton data. We use 2D skeletons as inputs and leverage a large human motion dataset for 2D to 3D pose estimation to learn generalizable motion features. A lightweight prompt generator produces instance-level prompts, and we employ dynamic queries with cross-attention to refine the semantic information of the input data. Additionally, we introduce a joint-enhanced multi-stream fusion mechanism based on self-attention to improve robustness against incomplete skeletons. Skeleton-Prompt represents a significant advancement in efficient fine-tuning for skeleton action recognition, effectively addressing cross-dataset generalization challenges in a data-efficient and parameter-efficient manner.",
author = "M. Lu and X. Lu and J. Liu",
year = "2025",
month = jun,
day = "9",
doi = "10.1016/j.patcog.2025.111885",
language = "English",
volume = "169",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Ltd",

}

RIS

TY - JOUR

T1 - Skeleton-prompt

T2 - A cross-dataset transfer learning approach for skeleton action recognition

AU - Lu, M.

AU - Lu, X.

AU - Liu, J.

PY - 2025/6/9

Y1 - 2025/6/9

N2 - This paper presents Skeleton-Prompt, a novel tuning method designed to tackle cross-dataset transfer issues in skeleton action recognition models. Given the scarcity of large-scale 3D skeleton datasets and the variability in keypoint structures across datasets, existing methods often rely on training models from scratch, necessitating extensive labeled data and exhibiting high sensitivity to occlusion. Our approach aims to fine-tune pre-trained models to adapt to limited real-world skeleton data. We use 2D skeletons as inputs and leverage a large human motion dataset for 2D to 3D pose estimation to learn generalizable motion features. A lightweight prompt generator produces instance-level prompts, and we employ dynamic queries with cross-attention to refine the semantic information of the input data. Additionally, we introduce a joint-enhanced multi-stream fusion mechanism based on self-attention to improve robustness against incomplete skeletons. Skeleton-Prompt represents a significant advancement in efficient fine-tuning for skeleton action recognition, effectively addressing cross-dataset generalization challenges in a data-efficient and parameter-efficient manner.

AB - This paper presents Skeleton-Prompt, a novel tuning method designed to tackle cross-dataset transfer issues in skeleton action recognition models. Given the scarcity of large-scale 3D skeleton datasets and the variability in keypoint structures across datasets, existing methods often rely on training models from scratch, necessitating extensive labeled data and exhibiting high sensitivity to occlusion. Our approach aims to fine-tune pre-trained models to adapt to limited real-world skeleton data. We use 2D skeletons as inputs and leverage a large human motion dataset for 2D to 3D pose estimation to learn generalizable motion features. A lightweight prompt generator produces instance-level prompts, and we employ dynamic queries with cross-attention to refine the semantic information of the input data. Additionally, we introduce a joint-enhanced multi-stream fusion mechanism based on self-attention to improve robustness against incomplete skeletons. Skeleton-Prompt represents a significant advancement in efficient fine-tuning for skeleton action recognition, effectively addressing cross-dataset generalization challenges in a data-efficient and parameter-efficient manner.

U2 - 10.1016/j.patcog.2025.111885

DO - 10.1016/j.patcog.2025.111885

M3 - Journal article

VL - 169

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

M1 - 111885

ER -