Neural Haircut: Prior-Guided Strand-Based Hair Reconstruction

TL;DR

Neural Haircut introduces a two-stage method for realistic, strand-level 3D hair reconstruction from monocular videos or multi-view images, combining volumetric modeling, learned priors, and differentiable rendering for high fidelity results.

Contribution

The paper presents a novel approach that integrates coarse volumetric reconstruction with strand-level refinement using learned hairstyle priors and image-based losses.

Findings

Achieves high realism and personalization in hair reconstruction.

Effective reconstruction from monocular videos under uncontrolled lighting.

Combines volumetric and strand-based modeling with differentiable rendering.

Abstract

Generating realistic human 3D reconstructions using image or video data is essential for various communication and entertainment applications. While existing methods achieved impressive results for body and facial regions, realistic hair modeling still remains challenging due to its high mechanical complexity. This work proposes an approach capable of accurate hair geometry reconstruction at a strand level from a monocular video or multi-view images captured in uncontrolled lighting conditions. Our method has two stages, with the first stage performing joint reconstruction of coarse hair and bust shapes and hair orientation using implicit volumetric representations. The second stage then estimates a strand-level hair reconstruction by reconciling in a single optimization process the coarse volumetric constraints with hair strand and hairstyle priors learned from the synthetic data. To further increase the reconstruction fidelity, we incorporate image-based losses into the fitting process using a new differentiable renderer. The combined system, named Neural Haircut, achieves high realism and personalization of the reconstructed hairstyles.