# Moulding Humans: Non-parametric 3D Human Shape Estimation from Single Images

## TL;DR

This paper introduces a non-parametric, double depth map approach for detailed 3D human shape estimation from single images, outperforming voxel-based methods in resolution and detail by using a depth-based representation and adversarial training.

## Contribution

It proposes a novel double depth map representation for high-resolution 3D human shape estimation from single images, incorporating adversarial training for improved realism.

## Key findings

- Achieves higher resolution 3D shape estimates than voxel-based methods.
- Demonstrates improved accuracy and 'humanness' through adversarial training.
- Validates approach on SURREAL and 3D-HUMANS datasets.

## Abstract

In this paper, we tackle the problem of 3D human shape estimation from single RGB images. While the recent progress in convolutional neural networks has allowed impressive results for 3D human pose estimation, estimating the full 3D shape of a person is still an open issue. Model-based approaches can output precise meshes of naked under-cloth human bodies but fail to estimate details and un-modelled elements such as hair or clothing. On the other hand, non-parametric volumetric approaches can potentially estimate complete shapes but, in practice, they are limited by the resolution of the output grid and cannot produce detailed estimates. In this work, we propose a non-parametric approach that employs a double depth map to represent the 3D shape of a person: a visible depth map and a "hidden" depth map are estimated and combined, to reconstruct the human 3D shape as done with a "mould". This representation through 2D depth maps allows a higher resolution output with a much lower dimension than voxel-based volumetric representations. Additionally, our fully derivable depth-based model allows us to efficiently incorporate a discriminator in an adversarial fashion to improve the accuracy and "humanness" of the 3D output. We train and quantitatively validate our approach on SURREAL and on 3D-HUMANS, a new photorealistic dataset made of semi-synthetic in-house videos annotated with 3D ground truth surfaces.