PartSDF: Part-Based Implicit Neural Representation for Composite 3D Shape Parametrization and Optimization

TL;DR

PartSDF introduces a part-based implicit neural representation for 3D shapes, enabling precise control and optimization of individual components within composite objects, improving reconstruction and generation in engineering tasks.

Contribution

It presents a novel supervised framework that explicitly models composite shapes with independent, controllable parts, enhancing shape manipulation and consistency.

Findings

Outperforms existing methods in shape reconstruction and generation

Enables precise control over individual shape components

Maintains overall shape coherence during part manipulation

Abstract

Accurate 3D shape representation is essential in engineering applications such as design, optimization, and simulation. In practice, engineering workflows require structured, part-based representations, as objects are inherently designed as assemblies of distinct components. However, most existing methods either model shapes holistically or decompose them without predefined part structures, limiting their applicability in real-world design tasks. We propose PartSDF, a supervised implicit representation framework that explicitly models composite shapes with independent, controllable parts while maintaining shape consistency. Thanks to its simple but innovative architecture, PartSDF outperforms both supervised and unsupervised baselines in reconstruction and generation tasks. We further demonstrate its effectiveness as a structured shape prior for engineering applications, enabling precise control over individual components while preserving overall coherence. Code available at https://github.com/cvlab-epfl/PartSDF.