Forecasting Human Dynamics From Static Images

Yu-Wei Chao, Jimei Yang, Brian Price, Scott Cohen, Jia Deng; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 548-556

Abstract


This paper presents the first study on forecasting human dynamics from static images. The problem is to input a single RGB image and generate a sequence of upcoming human body poses in 3D. To address the problem, we propose the 3D Pose Forecasting Network (3D-PFNet). Our 3D-PFNet integrates recent advances on single-image human pose estimation and sequence prediction, and converts the 2D predictions into 3D space. We train our 3D-PFNet using a three-step training strategy to leverage a diverse source of training data, including image and video based human pose datasets and 3D motion capture (MoCap) data. We demonstrate competitive performance of our 3D-PFNet on 2D pose forecasting and 3D structure recovery through quantitative and qualitative results.

Related Material


[pdf] [Supp]
[bibtex]
@InProceedings{Chao_2017_CVPR,
author = {Chao, Yu-Wei and Yang, Jimei and Price, Brian and Cohen, Scott and Deng, Jia},
title = {Forecasting Human Dynamics From Static Images},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {July},
year = {2017}
}