Scenes as Objects, Not Primitives: Instance-Structured 3D Tokenization from Unposed Views

Scenes as Objects, Not Primitives: Instance-Structured 3D Tokenization from Unposed Views

A feed-forward framework decomposes 3D scenes into instance-structured token groups from multi-view images, enabling direct object-level reconstruction, segmentation, and manipulat…

Hugging Face · Daily Papers ·Mijin Yoo, In Cho · ·▲ 34 upvotes

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

Autores: Mijin Yoo, In Cho, Subin Jeon, Jiwoo Lee, Eunbyung Park, Seon Joo Kim

  • 34 upvotes da comunidade
  • Temas: feed-forward framework, 3D scene decomposition, instance-structured token groups, unposed multi-view images, 3D Gaussians, differentiable rendering

Resumo

Resumo original (em inglês), extraído do paper:

A feed-forward framework decomposes 3D scenes into instance-structured token groups from multi-view images, enabling direct object-level reconstruction, segmentation, and manipulation without 3D annotations.

Onde ler

compartilhar: