State-of-the-art 3D human pose and shape estimation from video using GRUs and adversarial training.

Code:
arXiv:

Authors:
Muhammed Kocabas, Nikos Athanasiou, Michael J. Black

Abstract
Human motion is fundamental to understanding behavior. Despite progress on single-image 3D pose and shape estimation, existing video-based state-of-the-art methods fail to produce accurate and natural motion sequences due to a lack of ground-truth 3D motion data for training. To address this problem, we propose Video Inference for Body Pose and Shape Estimation (VIBE), which makes use of an existing large-scale motion capture dataset (AMASS, together with unpaired, in-the-wild, 2D keypoint annotations. Our key novelty is an adversarial learning framework that leverages AMASS to discriminate between real human motions and those produced by our temporal pose and shape regression networks. We define a temporal network architecture and show that adversarial training, at the sequence level, produces kinematically plausible motion sequences without in-the-wild ground-truth 3D labels. We perform extensive experimentation to analyze the importance of motion and demonstrate the effectiveness of VIBE on challenging 3D pose estimation datasets, achieving state-of-the-art performance.

Citation:
@inproceedings{VIBE:CVPR:2020,
title = {{VIBE}: Video Inference for Human Body Pose and Shape Estimation},
author = {Kocabas, Muhammed and Athanasiou, Nikos and Black, Michael J.},
booktitle = {Computer Vision and Pattern Recognition (CVPR)},
month = jun,
year = {2020}
}

source

0 Comments

Leave a reply

Think-Feel-Connect and NeverHateOnlyLove.com are registered trademarks of LOPROD®, all rights reserved. All our material, digital, written, audiovisual is property of our private network and its respective members, all rights are reserved. If you use our website in any form, it means you have accepted our Terms and Conditions. Questions? Contact us.


Think-Feel-Connect y NeverHateOnlyLove.com son marcas registradas de LOPROD®, todos los derechos reservados. Todo nuestro material, digital, escrito, audiovisual es propiedad de nuestra red privada y sus respectivos miembros, todos los derechos están reservados. Si utiliza nuestro sitio web en cualquier forma, significa que ha aceptado nuestros Términos y condiciones. ¿Preguntas? Contáctanos.

Log in with your credentials

Forgot your details?