Enhancing 3D Hand Pose Estimation via Dense Ordinal Regression Network


Yamin Mao (Samsung), Zhihua Liu (Samsung Research Center, Beijing), Weiming Li (Samsung), SoonYong Cho (Samsung), Qiang Wang (Samsung), Xiaoshuai Hao (Beijing Academy of Artificial Intelligence(BAAl) )
The 35th British Machine Vision Conference
PDFPosterVideo (Right click to download)

Abstract

Depth-based 3D hand pose estimation is an important but challenging task in robotics and autonomous driving. Recently, more attention has been given to dense regression methods for this task. These methods offer a good balance between accuracy and computational efficiency through the densely regressing hand joint offset maps. Despite the benefits, large-scale regression offset values are often affected by noise and outliers, leading to a significant drop in accuracy. To address this issue, we re-formulate 3D hand pose estimation as a dense ordinal regression problem and introduced a new Dense Ordinal Regression 3D Pose Network (DOR3D-Net). Specifically, we first decompose offset value regression into sub-tasks of binary classifications with ordinal constraints. Then, each binary classifier can predict the probability of a binary spatial relationship relative to joint, which is easier to train and yield much lower level of noise. The estimated hand joint positions are inferred by aggregating the ordinal regression results at local positions with a weighted sum. Furthermore, both joint regression loss and ordinal regression loss are used to train our DOR3D-Net in an end-to-end manner. Extensive experiments on public datasets (ICVL, MSRA, NYU and HANDS2017) show that our design provides significant improvements over SOTA methods.

Citation

@inproceedings{Mao_2024_BMVC,
author    = {Yamin Mao and Zhihua Liu and Weiming Li and SoonYong Cho and Qiang Wang and Xiaoshuai Hao},
title     = {Enhancing 3D Hand Pose Estimation via Dense Ordinal Regression Network},
booktitle = {35th British Machine Vision Conference 2024, {BMVC} 2024, Glasgow, UK, November 25-28, 2024},
publisher = {BMVA},
year      = {2024},
url       = {https://papers.bmvc2024.org/0245.pdf}
}


Copyright © 2024 The British Machine Vision Association and Society for Pattern Recognition
The British Machine Vision Conference is organised by The British Machine Vision Association and Society for Pattern Recognition. The Association is a Company limited by guarantee, No.2543446, and a non-profit-making body, registered in England and Wales as Charity No.1002307 (Registered Office: Dept. of Computer Science, Durham University, South Road, Durham, DH1 3LE, UK).

Imprint | Data Protection