English
 
User Manual Privacy Policy Disclaimer Contact us
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT

Released

Paper

XNect: Real-time Multi-person 3D Human Pose Estimation with a Single RGB Camera

MPS-Authors
/persons/resource/persons129023

Mehta,  Dushyant
Computer Graphics, MPI for Informatics, Max Planck Society;

/persons/resource/persons199773

Sotnychenko,  Oleksandr
Computer Graphics, MPI for Informatics, Max Planck Society;

/persons/resource/persons134216

Mueller,  Franziska
Computer Graphics, MPI for Informatics, Max Planck Society;

/persons/resource/persons206382

Xu,  Weipeng
Computer Graphics, MPI for Informatics, Max Planck Society;

/persons/resource/persons229949

Elgharib,  Mohamed
Computer Graphics, MPI for Informatics, Max Planck Society;

/persons/resource/persons45449

Seidel,  Hans-Peter
Computer Graphics, MPI for Informatics, Max Planck Society;

/persons/resource/persons118756

Pons-Moll,  Gerard
Computer Vision and Machine Learning, MPI for Informatics, Max Planck Society;

/persons/resource/persons45610

Theobalt,  Christian
Computer Graphics, MPI for Informatics, Max Planck Society;

Locator
There are no locators available
Fulltext (public)

arXiv:1907.00837.pdf
(Preprint), 10MB

Supplementary Material (public)
There is no public supplementary material available
Citation

Mehta, D., Sotnychenko, O., Mueller, F., Xu, W., Elgharib, M., Fua, P., et al. (2019). XNect: Real-time Multi-person 3D Human Pose Estimation with a Single RGB Camera. Retrieved from http://arxiv.org/abs/1907.00837.


Cite as: http://hdl.handle.net/21.11116/0000-0003-FE21-A
Abstract
We present a real-time approach for multi-person 3D motion capture at over 30 fps using a single RGB camera. It operates in generic scenes and is robust to difficult occlusions both by other people and objects. Our method operates in subsequent stages. The first stage is a convolutional neural network (CNN) that estimates 2D and 3D pose features along with identity assignments for all visible joints of all individuals. We contribute a new architecture for this CNN, called SelecSLS Net, that uses novel selective long and short range skip connections to improve the information flow allowing for a drastically faster network without compromising accuracy. In the second stage, a fully-connected neural network turns the possibly partial (on account of occlusion) 2D pose and 3D pose features for each subject into a complete 3D pose estimate per individual. The third stage applies space-time skeletal model fitting to the predicted 2D and 3D pose per subject to further reconcile the 2D and 3D pose, and enforce temporal coherence. Our method returns the full skeletal pose in joint angles for each subject. This is a further key distinction from previous work that neither extracted global body positions nor joint angle results of a coherent skeleton in real time for multi-person scenes. The proposed system runs on consumer hardware at a previously unseen speed of more than 30 fps given 512x320 images as input while achieving state-of-the-art accuracy, which we will demonstrate on a range of challenging real-world scenes.