Computer Graphics Laboratory ETH Zurich


How to Refine 3D Hand Pose Estimation from Unlabelled Depth Data ?

E. Dibra, T. Wolf, C. Öztireli, M. Gross

Proceedings of the Fifth International Conference on 3D Vision, 3DV (Qingdao, China, October 10-12, 2017), pp. 135-144


Data-driven approaches for hand pose estimation from depth images usually require a substantial amount of labelled training data which is quite hard to obtain. In this work, we show how a simple convolutional neural network, pre-trained only on synthetic depth images generated from a single 3D hand model, can be trained to adapt to unlabelled depth images from a real userís hand. We validate our method on two existing and a new dataset that we capture, both quantitatively and qualitatively, demonstrating that we strongly compare to state-of-the-art methods. Additionally, this method can be seen as an extension to existing methods trained on limited datasets, which helps on boosting their performance on new ones.


Download Paper
Download Paper
[PDF suppl.]
Download Video
Download Paper