English

Efficient, Self-Supervised Human Pose Estimation with Inductive Prior Tuning

Computer Vision and Pattern Recognition 2023-11-07 v1

Abstract

The goal of 2D human pose estimation (HPE) is to localize anatomical landmarks, given an image of a person in a pose. SOTA techniques make use of thousands of labeled figures (finetuning transformers or training deep CNNs), acquired using labor-intensive crowdsourcing. On the other hand, self-supervised methods re-frame the HPE task as a reconstruction problem, enabling them to leverage the vast amount of unlabeled visual data, though at the present cost of accuracy. In this work, we explore ways to improve self-supervised HPE. We (1) analyze the relationship between reconstruction quality and pose estimation accuracy, (2) develop a model pipeline that outperforms the baseline which inspired our work, using less than one-third the amount of training data, and (3) offer a new metric suitable for self-supervised settings that measures the consistency of predicted body part length proportions. We show that a combination of well-engineered reconstruction losses and inductive priors can help coordinate pose learning alongside reconstruction in a self-supervised paradigm.

Keywords

Cite

@article{arxiv.2311.02815,
  title  = {Efficient, Self-Supervised Human Pose Estimation with Inductive Prior Tuning},
  author = {Nobline Yoo and Olga Russakovsky},
  journal= {arXiv preprint arXiv:2311.02815},
  year   = {2023}
}

Comments

ICCVW 2023 Publication

R2 v1 2026-06-28T13:12:15.224Z