English

AniPixel: Towards Animatable Pixel-Aligned Human Avatar

Computer Vision and Pattern Recognition 2023-10-18 v2

Abstract

Although human reconstruction typically results in human-specific avatars, recent 3D scene reconstruction techniques utilizing pixel-aligned features show promise in generalizing to new scenes. Applying these techniques to human avatar reconstruction can result in a volumetric avatar with generalizability but limited animatability due to rendering only being possible for static representations. In this paper, we propose AniPixel, a novel animatable and generalizable human avatar reconstruction method that leverages pixel-aligned features for body geometry prediction and RGB color blending. Technically, to align the canonical space with the target space and the observation space, we propose a bidirectional neural skinning field based on skeleton-driven deformation to establish the target-to-canonical and canonical-to-observation correspondences. Then, we disentangle the canonical body geometry into a normalized neutral-sized body and a subject-specific residual for better generalizability. As the geometry and appearance are closely related, we introduce pixel-aligned features to facilitate the body geometry prediction and detailed surface normals to reinforce the RGB color blending. We also devise a pose-dependent and view direction-related shading module to represent the local illumination variance. Experiments show that AniPixel renders comparable novel views while delivering better novel pose animation results than state-of-the-art methods.

Keywords

Cite

@article{arxiv.2302.03397,
  title  = {AniPixel: Towards Animatable Pixel-Aligned Human Avatar},
  author = {Jinlong Fan and Jing Zhang and Zhi Hou and Dacheng Tao},
  journal= {arXiv preprint arXiv:2302.03397},
  year   = {2023}
}

Comments

Accepted by MM'23, code will be released at https://github.com/loong8888/AniPixel

R2 v1 2026-06-28T08:33:58.987Z