EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

THANet: Transferring Human Pose Estimation to Animal Pose Estimation

Jincheng Liao, Jianzhong Xu, Yunhang Shen, Shaohui Lin

Electronics · 11.10.2023 · 12 (20) · 4210

DOI: 10.3390/electronics12204210 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 11.10.2023. Verfasst von Jincheng Liao und 3 weiteren Autor:innen. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Shanghai Sailing Program, CCF-Tencent Open Research Fund.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Animal pose estimation (APE) boosts the understanding of animal behaviors. Recent vision-based APE has attracted extensive attention due to the advantages of contactless and sensorless applications. One of the main challenges in APE is the lack of high-quality keypoint annotations for different animal species since manually annotating the animal keypoints is very expensive and time-consuming. Existing works alleviate this problem by synthesizing APE data and generating pseudo-labels for unlabeled animal images. However, feature representations learned from synthetic images could not be directly transferred to real-world scenarios, and the generated pseudo-labels are usually noisy, which limits the model’s performance. To address the above challenge, we propose a novel cross-domain vision transformer for APE to Transfer Human pose estimation to Animal pose estimation, termed THANet, as humans share skeleton similarities with some animals. Inspired by the success of ViTPose in HPE, we design a unified vision transformer encoder to extract universal features for both animals and humans followed by two task-specific decoders. We further introduce a simple but effective cross-domain discriminator to bridge the domain gaps between the human pose and the animal pose. We evaluated the proposed THANet on the AP-10K and Animal-Pose benchmarks, and the extensive experiments show that our method achieves a promising performance. Specifically, the proposed vision transformer and cross-domain method significantly improve the model’s accuracy and generalization ability for APE.

Förderung

  • National Natural Science Foundation of China
  • Shanghai Sailing Program
  • CCF-Tencent Open Research Fund
  • Open Research Fund of Key Laboratory of Advanced Theory and Application in Statistics and Data Science, Ministry of Education
  • Fundamental Research Funds for the Central Universities

Zitieren

APA

Liao, J., Xu, J., Shen, Y., & Lin, S. (2023). THANet: Transferring Human Pose Estimation to Animal Pose Estimation. Electronics, 12(20), 4210. https://doi.org/10.3390/electronics12204210

Vancouver

Liao J, Xu J, Shen Y, Lin S. THANet: Transferring Human Pose Estimation to Animal Pose Estimation. Electronics. 2023;12(20):4210. doi:10.3390/electronics12204210

BibTeX
@article{Liao2023,
  author = {Liao, Jincheng and Xu, Jianzhong and Shen, Yunhang and Lin, Shaohui},
  title = {THANet: Transferring Human Pose Estimation to Animal Pose Estimation},
  journal = {Electronics},
  year = {2023},
  volume = {12},
  number = {20},
  pages = {4210},
  doi = {10.3390/electronics12204210}
}
Kostenlos starten →