End-to-end global to local convolutional neural network learning for hand pose recovery in depth data

Meysam Madadi*, Sergio Escalera, Xavier Baró, Jordi Gonzàlez

*Kontaktforfatter

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

1 Citationer (Scopus)
5 Downloads (Pure)

Abstract

Despite recent advances in 3-D pose estimation of human hands, thanks to the advent of convolutional neural networks (CNNs) and depth cameras, this task is still far from being solved in uncontrolled setups. This is mainly due to the highly non-linear dynamics of fingers and self-occlusions, which make hand model training a challenging task. In this study, a novel hierarchical tree-like structured CNN is exploited, in which branches are trained to become specialised in predefined subsets of hand joints called local poses. Further, local pose features, extracted from hierarchical CNN branches, are fused to learn higher order dependencies among joints in the final pose by end-to-end training. Lastly, the loss function used is also defined to incorporate appearance and physical constraints about doable hand motions and deformations. Finally, a non-rigid data augmentation approach is introduced to increase the amount of training depth data. Experimental results suggest that feeding a tree-shaped CNN, specialised in local poses, into a fusion network for modelling joints' correlations and dependencies, helps to increase the precision of final estimations, showing competitive results on NYU, MSRA, Hands17 and SyntheticHand datasets.

OriginalsprogEngelsk
TidsskriftIET Computer Vision
Vol/bind16
Udgave nummer1
Sider (fra-til)50-66
Antal sider17
ISSN1751-9632
DOI
StatusUdgivet - feb. 2022
Udgivet eksterntJa

Bibliografisk note

Publisher Copyright:
© 2021 The Authors. IET Computer Vision published by John Wiley & Sons Ltd on behalf of The Institution of Engineering and Technology.

Fingeraftryk

Dyk ned i forskningsemnerne om 'End-to-end global to local convolutional neural network learning for hand pose recovery in depth data'. Sammen danner de et unikt fingeraftryk.

Citationsformater