IGE-Net: Inverse Graphics Energy Networks for Human Pose Estimation and Single-View Reconstruction

Dominic Jack, Frederic Maire, Sareh Shirazi, Anders Eriksson; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 7075-7084

Abstract


Inferring 3D scene information from 2D observations is an open problem in computer vision. We propose using a deep-learning based energy minimization framework to learn a consistency measure between 2D observations and a proposed world model, and demonstrate that this framework can be trained end-to-end to produce consistent and realistic inferences. We evaluate the framework on human pose estimation and voxel-based object reconstruction benchmarks and show competitive results can be achieved with relatively shallow networks with drastically fewer learned parameters and floating point operations than conventional deep-learning approaches.

Related Material


[pdf] [supp]
[bibtex]
@InProceedings{Jack_2019_CVPR,
author = {Jack, Dominic and Maire, Frederic and Shirazi, Sareh and Eriksson, Anders},
title = {IGE-Net: Inverse Graphics Energy Networks for Human Pose Estimation and Single-View Reconstruction},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2019}
}