AMASS Dataset of 3D Human Motions

13 September 2019

03:22

Do you want to train deep networks to model 3D human pose and motion? Current mocap datasets are too small, so we created AMASS, a new dataset that unifies many mocap datasets by fitting the SMPL body model to them. AMASS has 45 hours of mocap and is growing. Contact us if you want to include your own mocap sequences in AMASS. Tutorial code shows how to use AMASS for deep learning. To appear: #ICCV2019 data: https://amass.is.tue.mpg.de/ code: https://github.com/nghorbani/amass pdf: http://files.is.tue.mpg.de/black/papers/amass.pdf @conference{AMASS:ICCV:2019, title = {{AMASS}: Archive of Motion Capture as Surface Shapes}, author = {Mahmood, Naureen and Ghorbani, Nima and Troje, Nikolaus F. and Pons-Moll, Gerard and Black, Michael J.}, booktitle = {International Conference on Computer Vision (ICCV)}, month = oct, year = {2019} }