DensiCrafter: Physically-Constrained Generation and Fabrication of Self-Supporting Hollow Structures

2025-11-12

AI Summary

This work addresses the challenge of generating lightweight, self-supporting, and manufacturable 3D hollow structures. We propose a novel differentiable density-field optimization framework. Its core innovation is a physics-free, differentiable self-support constraint—formulated as a geometric loss term—that jointly optimizes structural strength, weight, and printability while preserving geometric fidelity. The method initializes from voxel grids generated by Trellis, continuously embeds discrete structures into a differentiable density field, and integrates physical constraints—including mass regularization and domain-boundary constraints—into the loss function. It is compatible with pre-trained text-to-3D models such as Trellis and DSO. Experiments demonstrate that our optimized designs achieve up to 43% weight reduction over baselines, significantly improve mechanical stability and self-support capability, and—all validated via physical fabrication—successfully print without support structures.

Technology Category

• Search and Optimization: Learning to Search
• Computer Vision: Diffusion Models for Vision
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Sustainability and carbon-aware systems for Web, mobile, and WoT

Abstract

The rise of 3D generative models has enabled automatic 3D geometry and texture synthesis from multimodal inputs (e.g., text or images). However, these methods often ignore physical constraints and manufacturability considerations. In this work, we address the challenge of producing 3D designs that are both lightweight and self-supporting. We present DensiCrafter, a framework for generating lightweight, self-supporting 3D hollow structures by optimizing the density field. Starting from coarse voxel grids produced by Trellis, we interpret these as continuous density fields to optimize and introduce three differentiable, physically constrained, and simulation-free loss terms. Additionally, a mass regularization penalizes unnecessary material, while a restricted optimization domain preserves the outer surface. Our method seamlessly integrates with pretrained Trellis-based models (e.g., Trellis, DSO) without any architectural changes. In extensive evaluations, we achieve up to 43% reduction in material mass on the text-to-3D task. Compared to state-of-the-art baselines, our method could improve the stability and maintain high geometric fidelity. Real-world 3D-printing experiments confirm that our hollow designs can be reliably fabricated and could be self-supporting.

Problem

• Generating lightweight self-supporting 3D hollow structures
• Optimizing density fields under physical constraints
• Ensuring manufacturability while maintaining geometric fidelity

Innovation

• Optimizes density fields for hollow structures
• Uses differentiable physically-constrained loss terms
• Integrates with pretrained models without architectural changes