Prox-E: Fine-Grained 3D Shape Editing via Primitive-Based Abstractions

TL;DR

Prox-E introduces a training-free, primitive-based 3D shape editing framework that enables precise, localized modifications while maintaining overall shape identity, leveraging pretrained vision-language models for structural control.

Contribution

Prox-E is the first to combine primitive-based shape abstraction with pretrained vision-language models for fine-grained, training-free 3D shape editing.

Findings

Balances shape identity preservation and editing fidelity effectively.

Outperforms existing 2D-based and training-based 3D editing methods.

Enables localized structural modifications without retraining.

Abstract

Text-based 2D image editing models have recently reached an impressive level of maturity, motivating a growing body of work that heavily depends on these models to drive 3D edits. While effective for appearance-based modifications, such 2D-centric 3D editing pipelines often struggle with fine-grained 3D editing, where localized structural changes must be applied while strictly preserving an object's overall identity. To address this limitation, we propose Prox-E, a training-free framework that enables fine-grained 3D control through an explicit, primitive-based geometric abstraction. Our framework first abstracts an input 3D shape into a compact set of geometric primitives. A pretrained vision-language model (VLM) then edits this abstraction to specify primitive-level changes. These structural edits are subsequently used to guide a 3D generative model, enabling fine-grained, localized modifications while preserving unchanged regions of the original shape. Through extensive experiments, we demonstrate that our method consistently balances identity preservation, shape quality, and instruction fidelity more effectively than various existing approaches, including 2D-based 3D editors and training-based methods.