SpatialAvatar-0: High-Quality 4D Head Avatar with Multi-Stage Reconstruction

SpatialAvatar-0: High-Quality 4D Head Avatar with Multi-Stage Reconstruction

SpatialAvatar-0 enables high-quality 4D head avatar generation by combining feed-forward prediction with per-subject refinement through a shared Gaussian representation, achieving…

Hugging Face · Daily Papers ·Yiran Wang, Zeyu Zhang · ·▲ 3 upvotes

Este artigo está em destaque na seleção diária de papers do Hugging Face, curada pela comunidade de pesquisa em IA.

Autores: Yiran Wang, Zeyu Zhang, Yuanming Li, Ziming Wang, Yang Zhao

  • 3 upvotes da comunidade
  • Temas: 3D Gaussian Splatting, FLAME-mesh-bound Gaussian representation, feed-forward predictor, per-subject refiner, mean-pool, monocular-temporal to multi-view-spatial scheduling

Resumo

Resumo original (em inglês), extraído do paper:

SpatialAvatar-0 enables high-quality 4D head avatar generation by combining feed-forward prediction with per-subject refinement through a shared Gaussian representation, achieving superior performance across multiple benchmarks.

Ler o paper completo no Hugging Face →

compartilhar: