3D Convolutional Neural Networks for Efficient and Robust
Hand Pose Estimation from Single Depth Images



Authors
    Liuhao Ge
    Hui Liang
    Junsong Yuan

Abstract
We propose a simple, yet effective approach for real-time hand pose estimation from single depth images using three-dimensional Convolutional Neural Networks (3D CNNs). Image based features extracted by 2D CNNs are not directly suitable for 3D hand pose estimation due to the lack of 3D spatial information. Our proposed 3DCNN taking a 3D volumetric representation of the hand depth image as input can capture the 3D spatial structure of the input and accurately regress full 3D hand pose in a single pass. In order to make the 3D CNN robust to variations in hand sizes and global orientations, we perform 3D data augmentation on the training data. Experiments show that our proposed 3D CNN based approach outperforms state-of-the-art methods on two challenging hand pose datasets, and is very efficient as our implementation runs at over 215 fps on a standard computer with a single GPU.

Paper
Liuhao Ge, Hui Liang, Junsong Yuan and Daniel Thalmann, "Real-time 3D Hand Pose Estimation with 3D Convolutional Neural Networks", IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018. Accepted.

Liuhao Ge, Hui Liang, Junsong Yuan and Daniel Thalmann, "3D Convolutional Neural Networks for Efficient and Robust Hand Pose Estimation from Single Depth Images", Proceedings of CVPR 2017. [pdf] [bibtex] [video] [result]