Reposing Humans by Warping 3D Features

TL;DR

This paper introduces a novel method for human reposing by learning a dense volumetric feature representation that can be explicitly warped to generate images in new poses, outperforming previous approaches.

Contribution

It proposes a deep learning-based volumetric feature volume for human reposing, enabling intuitive geometric warping without explicit 3D mesh fitting.

Findings

Achieves state-of-the-art results on DeepFashion and iPER benchmarks.

Demonstrates effective manipulation of 3D human features through simple warping.

Outperforms prior 2D and mesh-based methods in pose transfer quality.

Abstract

We address the problem of reposing an image of a human into any desired novel pose. This conditional image-generation task requires reasoning about the 3D structure of the human, including self-occluded body parts. Most prior works are either based on 2D representations or require fitting and manipulating an explicit 3D body mesh. Based on the recent success in deep learning-based volumetric representations, we propose to implicitly learn a dense feature volume from human images, which lends itself to simple and intuitive manipulation through explicit geometric warping. Once the latent feature volume is warped according to the desired pose change, the volume is mapped back to RGB space by a convolutional decoder. Our state-of-the-art results on the DeepFashion and the iPER benchmarks indicate that dense volumetric human representations are worth investigating in more detail.