Fast Texture Transfer for XR Avatars via Barycentric UV Conversion

2025-08-26

AI Summary

To address the inefficiency and severe boundary artifacts inherent in conventional affine transformations for facial texture transfer onto SMPL-X avatars, this paper proposes an efficient UV coordinate transformation method based on barycentric coordinates. By precomputing a global UV mapping as a single linear transformation matrix, the approach enables one-shot, artifact-free synthesis of source facial textures directly into the target SMPL-X facial UV domain. This eliminates per-pixel affine computations, reducing texture transfer to a matrix multiplication followed by bilinear sampling. Experiments demonstrate that our method achieves over 7,000× faster inference compared to baseline approaches, while significantly improving texture continuity and visual fidelity. Quantitative evaluations (PSNR/SSIM) and qualitative assessments in XR scenarios confirm its effectiveness and practicality for real-time personalized avatar generation.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Cognitive Modeling & Cognitive Systems: Affective Computing
• Machine Learning: Learning with Manifolds

Application Category

• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

We present a fast and efficient method for transferring facial textures onto SMPL-X-based full-body avatars. Unlike conventional affine-transform methods that are slow and prone to visual artifacts, our method utilizes a barycentric UV conversion technique. Our approach precomputes the entire UV mapping into a single transformation matrix, enabling texture transfer in a single operation. This results in a speedup of over 7000x compared to the baseline, while also significantly improving the final texture quality by eliminating boundary artifacts. Through quantitative and qualitative evaluations, we demonstrate that our method offers a practical solution for personalization in immersive XR applications. The code is available online.

Problem

• Fast facial texture transfer for XR avatars
• Eliminating boundary artifacts in texture mapping
• Accelerating texture transfer via barycentric UV conversion

Innovation

• Barycentric UV conversion technique
• Precomputed single transformation matrix
• 7000x speedup eliminating boundary artifacts