OmniPart: Part-Aware 3D Generation with Semantic Decoupling and Structural Cohesion

TL;DR

OmniPart introduces a part-aware 3D generation framework that decouples structure planning from detailed synthesis, enabling controllable, editable, and diverse 3D asset creation with high semantic and structural fidelity.

Contribution

The paper presents a novel two-stage approach combining structure planning and spatially-conditioned synthesis, improving controllability and interpretability in 3D generative models.

Findings

Achieves state-of-the-art performance in part-aware 3D generation

Supports user-defined part granularity and precise localization

Enables diverse and editable 3D content creation

Abstract

The creation of 3D assets with explicit, editable part structures is crucial for advancing interactive applications, yet most generative methods produce only monolithic shapes, limiting their utility. We introduce OmniPart, a novel framework for part-aware 3D object generation designed to achieve high semantic decoupling among components while maintaining robust structural cohesion. OmniPart uniquely decouples this complex task into two synergistic stages: (1) an autoregressive structure planning module generates a controllable, variable-length sequence of 3D part bounding boxes, critically guided by flexible 2D part masks that allow for intuitive control over part decomposition without requiring direct correspondences or semantic labels; and (2) a spatially-conditioned rectified flow model, efficiently adapted from a pre-trained holistic 3D generator, synthesizes all 3D parts simultaneously and consistently within the planned layout. Our approach supports user-defined part granularity, precise localization, and enables diverse downstream applications. Extensive experiments demonstrate that OmniPart achieves state-of-the-art performance, paving the way for more interpretable, editable, and versatile 3D content.