Projects per year
Abstract
Crucial to the success of training a depth-based 3D hand pose estimator (HPE) is the availability of comprehensive datasets covering diverse camera perspectives, shapes, and pose variations. However, collecting such annotated datasets is challenging. We propose to complete existing databases by generating new database entries. The key idea is to synthesize data in the skeleton space (instead of doing so in the depth-map space) which enables an easy and intuitive way of manipulating data entries. Since the skeleton entries generated in this way do not have the corresponding depth map entries, we exploit them by training a separate hand pose generator (HPG) which synthesizes the depth map from the skeleton entries. By training the HPG and HPE in a single unified optimization framework enforcing that 1) the HPE agrees with the paired depth and skeleton entries; and 2) the HPG-HPE combination satisfies the cyclic consistency (both the input and the output of HPG-HPE are skeletons) observed via the newly generated unpaired skeletons, our algorithm constructs a HPE which is robust to variations that go beyond the coverage of the existing database.
Our training algorithm adopts the generative adversarial networks (GAN) training process. As a by-product, we obtain a hand pose discriminator (HPD) that is capable of picking out realistic hand poses. Our algorithm exploits this capability to refine the initial skeleton estimates in testing, further improving the accuracy. We test our algorithm on four challenging benchmark datasets (ICVL, MSRA, NYU and Big Hand 2.2M datasets) and demonstrate that our approach outperforms or is on par with state-of-the-art methods quantitatively and qualitatively.
Our training algorithm adopts the generative adversarial networks (GAN) training process. As a by-product, we obtain a hand pose discriminator (HPD) that is capable of picking out realistic hand poses. Our algorithm exploits this capability to refine the initial skeleton estimates in testing, further improving the accuracy. We test our algorithm on four challenging benchmark datasets (ICVL, MSRA, NYU and Big Hand 2.2M datasets) and demonstrate that our approach outperforms or is on par with state-of-the-art methods quantitatively and qualitatively.
Original language | English |
---|---|
Title of host publication | 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition |
Publisher | IEEE |
Pages | 8330-8339 |
Number of pages | 10 |
Volume | 2018 |
ISBN (Electronic) | 9781538664209 |
ISBN (Print) | 9781538664216 |
DOIs | |
Publication status | Published - 17 Dec 2018 |
Event | IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2018 - Duration: 18 Jun 2018 → 22 Jun 2018 |
Publication series
Name | Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition) |
---|---|
Publisher | IEEE |
ISSN (Print) | 2575-7075 |
Conference
Conference | IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2018 |
---|---|
Period | 18/06/18 → 22/06/18 |
Fingerprint
Dive into the research topics of 'Augmented Skeleton Space Transfer for Depth-based Hand Pose Estimation'. Together they form a unique fingerprint.Projects
- 2 Finished
-
Personalized Exploration of Imagery Database
Kim, K. I. (PI)
Engineering and Physical Sciences Research Council
1/09/16 → 31/05/17
Project: Research council
-
Centre for the Analysis of Motion, Entertainment Research and Applications (CAMERA)
Cosker, D. (PI), Bilzon, J. (CoI), Campbell, N. (CoI), Cazzola, D. (CoI), Colyer, S. (CoI), Fincham Haines, T. (CoI), Hall, P. (CoI), Kim, K. I. (CoI), Lutteroth, C. (CoI), McGuigan, P. (CoI), O'Neill, E. (CoI), Richardt, C. (CoI), Salo, A. (CoI), Seminati, E. (CoI), Tabor, A. (CoI) & Yang, Y. (CoI)
Engineering and Physical Sciences Research Council
1/09/15 → 28/02/21
Project: Research council