PrimitiveAnything: Human-Crafted 3D Primitive Assembly Generation with Auto-Regressive Transformer
Jingwen Ye, Yuze He, Yanning Zhou, Yiqin Zhu, Kaiwen Xiao, Yong-Jin, Liu, Wei Yang, Xiao Han

TL;DR
PrimitiveAnything is a novel auto-regressive transformer framework that learns to generate human-like 3D primitive assemblies from large-scale data, improving shape abstraction and generalization across diverse categories.
Contribution
It reformulates shape primitive abstraction as an assembly generation task using a shape-conditioned transformer and a unified primitive parameterization scheme, enabling better generalization and human-aligned decomposition.
Findings
Generates high-quality primitive assemblies aligned with human perception
Maintains geometric fidelity across diverse shape categories
Enables applications in 3D content creation and user-generated content
Abstract
Shape primitive abstraction, which decomposes complex 3D shapes into simple geometric elements, plays a crucial role in human visual cognition and has broad applications in computer vision and graphics. While recent advances in 3D content generation have shown remarkable progress, existing primitive abstraction methods either rely on geometric optimization with limited semantic understanding or learn from small-scale, category-specific datasets, struggling to generalize across diverse shape categories. We present PrimitiveAnything, a novel framework that reformulates shape primitive abstraction as a primitive assembly generation task. PrimitiveAnything includes a shape-conditioned primitive transformer for auto-regressive generation and an ambiguity-free parameterization scheme to represent multiple types of primitives in a unified manner. The proposed framework directly learns the…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
Topics3D Shape Modeling and Analysis · Human Motion and Animation · Robot Manipulation and Learning
MethodsALIGN
