Monocular real-time hand shape and motion capture using multi-modal data
We present a novel method for monocular hand shape and pose estimation at
unprecedented runtime performance of 100fps and at state-of-the-art accuracy. This is
enabled by a new learning based architecture designed such that it can make use of all the
sources of available hand training data: image data with either 2D or 3D annotations, as well
as stand-alone 3D animations without corresponding image data. It features a 3D hand joint
detection module and an inverse kinematics module which regresses not only 3D joint …
unprecedented runtime performance of 100fps and at state-of-the-art accuracy. This is
enabled by a new learning based architecture designed such that it can make use of all the
sources of available hand training data: image data with either 2D or 3D annotations, as well
as stand-alone 3D animations without corresponding image data. It features a 3D hand joint
detection module and an inverse kinematics module which regresses not only 3D joint …
[PDF][PDF] Monocular Real-time Hand Shape and Motion Capture using Multi-modal Data–Supplemental Document–
YZM Habermann, W Xu, I Habibie, C Theobalt, F Xu - openaccess.thecvf.com
In the following, we provide more quantitative evaluations and comparisons to previous
methods (Sec. 1). Further, we show more qualitative results of our method on sequences
from the internet and publicly available datasets (Sec. 2). Finally, we provide more technical
details about our method (Sec. 3).
methods (Sec. 1). Further, we show more qualitative results of our method on sequences
from the internet and publicly available datasets (Sec. 2). Finally, we provide more technical
details about our method (Sec. 3).
Showing the best results for this search. See all results