arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2003.04030cs.CV
收藏

多人姿态估计的精细局部表示学习

Learning Delicate Local Representations for Multi-Person Pose Estimation

Tsinghua University(清华大学);Chinese Academy of Sciences(中国科学院);Beihang University(北京航空航天大学);Ocean University of China(中国海洋大学);Megvii Inc.(旷视科技有限公司)

Yuanhao Cai, Zhicheng Wang, Zhengxiong Luo, Binyi Yin, Angang Du, Haoqian Wang, Xiangyu Zhang, Xinyu Zhou, Erjin Zhou, Jian Sun

摘要:

在本文中,我们提出了一种新的方法称为剩余步骤网络(RSN)。RSN有效地聚合具有相同空间大小的特征(层内特征)以获得精细的局部表示,其保留丰富的低层空间信息并导致精确的关键点定位。此外,我们观察到输出功能对最终性能的贡献不同。为了解决这个问题,我们提出了一个有效的注意力机制-姿态优化机(PRM),在输出特征的局部和全局表示之间进行权衡,并进一步细化关键点位置。我们的方法赢得了COCO Keypoint Challenge 2019的第一名,并在COCO和MPII基准测试中取得了最先进的结果,而无需使用额外的训练数据和预训练模型。我们的单个模型在COCO test-dev上获得78.6,在MPII测试数据集上获得93.0。集成模型在COCO测试-开发数据集上获得79.2,在COCO测试-挑战数据集上获得77.1。源代码可在https://github.com/caiyuanhao1998/RSN/上公开获取,以供进一步研究

英文摘要:

In this paper, we propose a novel method called Residual Steps Network (RSN). RSN aggregates features with the same spatial size (Intra-level features) efficiently to obtain delicate local representations, which retain rich low-level spatial information and result in precise keypoint localization. Additionally, we observe the output features contribute differently to final performance. To tackle this problem, we propose an efficient attention mechanism - Pose Refine Machine (PRM) to make a trade-off between local and global representations in output features and further refine the keypoint locations. Our approach won the 1st place of COCO Keypoint Challenge 2019 and achieves state-of-the-art results on both COCO and MPII benchmarks, without using extra training data and pretrained model. Our single model achieves 78.6 on COCO test-dev, 93.0 on MPII test dataset. Ensembled models achieve 79.2 on COCO test-dev, 77.1 on COCO test-challenge dataset. The source code is publicly available for further research at https://github.com/caiyuanhao1998/RSN/

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑