Prox-E is a training-free pipeline that abstracts 3D shapes into editable geometric primitives, uses a VLM to specify changes, and guides a generative model to apply precise local edits.
William Gao, Noam Aigerman, Thibault Groueix, Vova Kim, and Rana Hanocka
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
C3G creates compact 3D Gaussian representations with 2K points by guiding placement via learnable tokens that aggregate multi-view features through attention, yielding better efficiency and performance than dense methods.
citing papers explorer
-
Prox-E: Fine-Grained 3D Shape Editing via Primitive-Based Abstractions
Prox-E is a training-free pipeline that abstracts 3D shapes into editable geometric primitives, uses a VLM to specify changes, and guides a generative model to apply precise local edits.
-
C3G: Learning Compact 3D Representations with 2K Gaussians
C3G creates compact 3D Gaussian representations with 2K points by guiding placement via learnable tokens that aggregate multi-view features through attention, yielding better efficiency and performance than dense methods.