HandMap: Robust Hand Pose Estimation via Intermediate Dense Guidance Map Supervision

Xiaokun Wu, Daniel Finnegan, Eamonn O'Neill, Yongliang Yang

Research output: Chapter or section in a book/report/conference proceedingChapter in a published conference proceeding

4 Citations (SciVal)
70 Downloads (Pure)

Abstract

This work presents a novel hand pose estimation framework via intermediate dense guidance map supervision. By leveraging the advantage of predicting heat maps of hand joints in detection-based methods, we propose to use dense feature maps through intermediate supervision in a regression-based framework that is not limited to the resolution of the heat map. Our dense feature maps are delicately designed to encode the hand geometry and the spatial relation between local joint and global hand. The proposed framework significantly improves the state-of-the-art in both 2D and 3D on the recent benchmark datasets.
Original languageEnglish
Title of host publicationComputer Vision – ECCV 2018 - 15th European Conference, 2018, Proceedings
Subtitle of host publicationComputer Vision – ECCV 2018
EditorsVittorio Ferrari, Martial Hebert, Christian Sminchisescu, Yair Weiss
PublisherSpringer
Pages246-262
Number of pages17
ISBN (Electronic)978-3-030-01270-0
ISBN (Print)978-3-030-01269-4
DOIs
Publication statusPublished - 2018
Event15th European Conference on Computer Vision - Munich, Germany
Duration: 10 Sept 201813 Sept 2018

Publication series

Name Lecture Notes in Computer Science
PublisherSpringer
Volume11220
NameImage Processing, Computer Vision, Pattern Recognition, and Graphics
Volume11220

Conference

Conference15th European Conference on Computer Vision
Country/TerritoryGermany
CityMunich
Period10/09/1813/09/18

Keywords

  • Dense guidance map
  • Hand pose estimation
  • Intermediate supervision

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'HandMap: Robust Hand Pose Estimation via Intermediate Dense Guidance Map Supervision'. Together they form a unique fingerprint.

Cite this