EMDB: The Electromagnetic Database of Global 3D Human Pose and Shape in the Wild

TL;DR

EMDB is a new in-the-wild dataset of 3D human pose and shape, captured with electromagnetic sensors and iPhone, providing high-quality global trajectories and detailed surface geometry for advancing pose estimation research.

Contribution

The paper introduces EMDB, a comprehensive dataset with global 3D human pose, shape, and trajectories, created using a novel multi-stage optimization and neural implicit modeling.

Findings

EMDB achieves 2.3 cm positional accuracy and 10.6° angular error.

State-of-the-art monocular methods are evaluated on EMDB.

EMDB surpasses previous in-the-wild datasets in accuracy.

Abstract

We present EMDB, the Electromagnetic Database of Global 3D Human Pose and Shape in the Wild. EMDB is a novel dataset that contains high-quality 3D SMPL pose and shape parameters with global body and camera trajectories for in-the-wild videos. We use body-worn, wireless electromagnetic (EM) sensors and a hand-held iPhone to record a total of 58 minutes of motion data, distributed over 81 indoor and outdoor sequences and 10 participants. Together with accurate body poses and shapes, we also provide global camera poses and body root trajectories. To construct EMDB, we propose a multi-stage optimization procedure, which first fits SMPL to the 6-DoF EM measurements and then refines the poses via image observations. To achieve high-quality results, we leverage a neural implicit avatar model to reconstruct detailed human surface geometry and appearance, which allows for improved alignment and smoothness via a dense pixel-level objective. Our evaluations, conducted with a multi-view volumetric capture system, indicate that EMDB has an expected accuracy of 2.3 cm positional and 10.6 degrees angular error, surpassing the accuracy of previous in-the-wild datasets. We evaluate existing state-of-the-art monocular RGB methods for camera-relative and global pose estimation on EMDB. EMDB is publicly available under https://ait.ethz.ch/emdb