Refining 3D Human Texture Estimation from a Single Image

TL;DR

This paper introduces a novel framework for 3D human texture estimation from a single image, utilizing adaptive sampling, cycle consistency, and uncertainty-based loss to improve quality and view generalization.

Contribution

It presents a new method combining deformable convolution, cycle consistency, and uncertainty-based loss for superior 3D human texture estimation from single images.

Findings

Significant qualitative improvements over state-of-the-art methods.

Quantitative results demonstrate higher accuracy in texture reconstruction.

Enhanced view generalization through cycle consistency loss.

Abstract

Estimating 3D human texture from a single image is essential in graphics and vision. It requires learning a mapping function from input images of humans with diverse poses into the parametric (UV) space and reasonably hallucinating invisible parts. To achieve a high-quality 3D human texture estimation, we propose a framework that adaptively samples the input by a deformable convolution where offsets are learned via a deep neural network. Additionally, we describe a novel cycle consistency loss that improves view generalization. We further propose to train our framework with an uncertainty-based pixel-level image reconstruction loss, which enhances color fidelity. We compare our method against the state-of-the-art approaches and show significant qualitative and quantitative improvements.