Abstract

Multi-person pose estimation in the wild is challenging.Although state-of-the-art human detectors have demonstrated good performance, small errors in localization and recognition are inevitable. These errors can cause failures for a single-person pose estimator (SPPE), especially for methods that solely depend on human detection results. In this paper, we propose a novel regional multi-person pose estimation (RMPE) framework to facilitate pose estimation in the presence of inaccurate human bounding boxes. Our framework consists of three components: Symmetric Spatial Transformer Network (SSTN), Parametric Pose NonMaximum-Suppression (NMS), and Pose-Guided Proposals Generator (PGPG). Our method is able to handle inaccurate bounding boxes and redundant detections, allowing it to achieve 17% increase on the MPII (multi-person) dataset.

Authors


Hao-Shu Fang
Shuqin Xie
Yu-Wing Tai
Cewu Lu

Demo

Code

Our source code is available on Github, including:

  • Training/test code
  • Pretrained model
  • Evaluation code

Paper

Download the

paper here.

Bibtex

@inproceedings{fang2017rmpe,
                   title={RMPE: Regional Multi-person Pose Estimation},
                   author={Fang, Hao-Shu and Xie, Shuqin and Tai, Yu-Wing and Lu, Cewu},
                   booktitle={ICCV},
                   year={2017}
                  }

Example Results


Browse more results in the supplementary material or use our github code.