Resolution irrelevant encoding and difficulty balanced loss based network independent supervision for multi-person pose estimation

Haiyang Liu, Dingli Luo, Songlin Du, Takeshi Ikenaga

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Sustainable efforts are made to improve the accuracy performance in multi-person pose estimation, but the current accuracy is still not enough for real-world applications. Besides, most improvement approaches are designed for special basement networks and ignore the speed performance, which results in limited applicability and low cost-performance. This paper proposes two network independent supervision: Resolution Irrelevant Encoding and Difficulty Balanced Loss. The proposed methods reorganize task representatives, the loss calculation method, and the loss punishment ratio in one-stage pose estimation frameworks to improve the joints' location accuracy with general applicability and high computational efficiency. Resolution Irrelevant Encoding fuses heatmaps and proposed inner block offsets to fix pixel-level joints positions without resolution limitations. To improve network training efficiency, Difficulty Balanced Loss adjusts loss weight in spatial and sequential aspects. On the MS COCO keypoints detection benchmark, the mAP of OpenPose trained with our proposals outperforms the OpenPose baseline over 4.9%.

Original languageEnglish
Title of host publicationProceedings - 2020 13th International Conference on Human System Interaction, HSI 2020
PublisherIEEE Computer Society
Pages112-117
Number of pages6
ISBN (Electronic)9781728173924
DOIs
Publication statusPublished - 2020 Jun
Event13th International Conference on Human System Interaction, HSI 2020 - Tokyo, Japan
Duration: 2020 Jun 62020 Jun 8

Publication series

NameInternational Conference on Human System Interaction, HSI
Volume2020-June
ISSN (Print)2158-2246
ISSN (Electronic)2158-2254

Conference

Conference13th International Conference on Human System Interaction, HSI 2020
CountryJapan
CityTokyo
Period20/6/620/6/8

Keywords

  • Human pose estimation
  • Network independent
  • Supervision strategy

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Software

Fingerprint Dive into the research topics of 'Resolution irrelevant encoding and difficulty balanced loss based network independent supervision for multi-person pose estimation'. Together they form a unique fingerprint.

Cite this