Unite the People: Closing the Loop Between 3D and 2D Human Representations

TL;DR

This paper introduces a hybrid approach combining 3D modeling and 2D estimation to create a large, richly annotated dataset for human body analysis, enabling state-of-the-art 3D pose and shape estimation with less data.

Contribution

It presents UP-3D, a new dataset created using an extended SMPLify method with human sorting, and demonstrates how this data improves detailed 3D human pose and shape estimation.

Findings

Achieved state-of-the-art 3D human pose estimation results.

Predicted 31 body segments and 91 landmarks with high accuracy.

Reduced training data requirements by an order of magnitude.

Abstract

3D models provide a common ground for different representations of human bodies. In turn, robust 2D estimation has proven to be a powerful tool to obtain 3D fits "in-the- wild". However, depending on the level of detail, it can be hard to impossible to acquire labeled data for training 2D estimators on large scale. We propose a hybrid approach to this problem: with an extended version of the recently introduced SMPLify method, we obtain high quality 3D body model fits for multiple human pose datasets. Human annotators solely sort good and bad fits. This procedure leads to an initial dataset, UP-3D, with rich annotations. With a comprehensive set of experiments, we show how this data can be used to train discriminative models that produce results with an unprecedented level of detail: our models predict 31 segments and 91 landmark locations on the body. Using the 91 landmark pose estimator, we present state-of-the art results for 3D human pose and shape estimation using an order of magnitude less training data and without assumptions about gender or pose in the fitting procedure. We show that UP-3D can be enhanced with these improved fits to grow in quantity and quality, which makes the system deployable on large scale. The data, code and models are available for research purposes.