Affordable Access

Access to the full text

Depth-Based Hand Pose Estimation: Methods, Data, and Challenges

Authors
  • Supančič, James Steven III1
  • Rogez, Grégory2, 3
  • Yang, Yi4
  • Shotton, Jamie5
  • Ramanan, Deva6
  • 1 University of California, Irvine, USA , Irvine (United States)
  • 2 Univ. Grenoble Alpes, Inria, CNRS, Grenoble INP*, LJK, Grenoble, 38000, France , Grenoble (France)
  • 3 Institute of Engineering Univ., Grenoble Alpes, France , Grenoble Alpes (France)
  • 4 Baidu Institute of Deep Learning, Sunnyvale, USA , Sunnyvale (United States)
  • 5 Microsoft Research, Cambridge, UK , Cambridge (United Kingdom)
  • 6 Carnegie Mellon University, Pittsburgh, PA, USA , Pittsburgh (United States)
Type
Published Article
Journal
International Journal of Computer Vision
Publisher
Springer-Verlag
Publication Date
Apr 12, 2018
Volume
126
Issue
11
Pages
1180–1198
Identifiers
DOI: 10.1007/s11263-018-1081-7
Source
Springer Nature
Keywords
License
Yellow

Abstract

Hand pose estimation has matured rapidly in recent years. The introduction of commodity depth sensors and a multitude of practical applications have spurred new advances. We provide an extensive analysis of the state-of-the-art, focusing on hand pose estimation from a single depth frame. To do so, we have implemented a considerable number of systems, and have released software and evaluation code. We summarize important conclusions here: (1) Coarse pose estimation appears viable for scenes with isolated hands. However, high precision pose estimation [required for immersive virtual reality and cluttered scenes (where hands may be interacting with nearby objects and surfaces) remain a challenge. To spur further progress we introduce a challenging new dataset with diverse, cluttered scenes. (2) Many methods evaluate themselves with disparate criteria, making comparisons difficult. We define a consistent evaluation criteria, rigorously motivated by human experiments. (3) We introduce a simple nearest-neighbor baseline that outperforms most existing systems. This implies that most systems do not generalize beyond their training sets. This also reinforces the under-appreciated point that training data is as important as the model itself. We conclude with directions for future progress.

Report this publication

Statistics

Seen <100 times