DreamEdit3D: Personalization of Multi-View Diffusion Models for 3D Editing

TL;DR

DreamEdit3D introduces a novel method for personalized, text-guided 3D editing that achieves multi-view consistency and object-level control by learning disentangled token embeddings for 3D assets.

Contribution

It proposes a two-phase optimization strategy for 3D personalization, enabling compositional control and high-fidelity 3D mesh generation from multi-view images.

Findings

Achieves state-of-the-art edit faithfulness.

Maintains identity preservation across edits.

Enables object-level, natural language control in 3D editing.

Abstract

While 2D diffusion models have achieved remarkable success in identity-preserving personalization, extending this capability to 3D assets remains a significant challenge due to the complexities of multi-view consistency and spatial control. Inspired by these 2D advancements, we present a novel personalization method for text-guided 3D editing that enables compositional, object-level control through natural language. Given a 3D input, we render orthogonal views and extract object-level segmentation masks to isolate semantic components. We then learn distinct token embeddings for each component through a tailored two-phase optimization strategy: multi-view textual inversion with attention alignment, followed by full fine-tuning of multi-view diffusion model. During inference, these disentangled tokens seamlessly compose with editing prompts to generate multi-view consistent images, which are subsequently lifted into high-fidelity textured 3D meshes. Extensive evaluations across diverse editing scenarios demonstrate that our method successfully transfers the flexibility of 2D personalization to 3D, achieving state-of-the-art edit faithfulness and identity preservation compared to existing baselines.