Chirpy3D: Creative Fine-grained 3D Object Fabrication via Part Sampling

TL;DR

Chirpy3D introduces a zero-shot 3D object generation method that synthesizes creative, fine-grained 3D models by unsupervised part sampling and composition, without requiring detailed annotations.

Contribution

It presents the first system capable of generating novel 3D objects with detailed species-specific features through unsupervised part decomposition and flexible recombination.

Findings

Outperforms existing methods in quality and detail of generated 3D objects.

Enables smooth interpolation and recombination of parts for new object creation.

Achieves zero-shot generalization to unseen categories.

Abstract

We present Chirpy3D, a novel approach for fine-grained 3D object generation, tackling the challenging task of synthesizing creative 3D objects in a zero-shot setting, with access only to unposed 2D images of seen categories. Without structured supervision -- such as camera poses, 3D part annotations, or object-specific labels -- the model must infer plausible 3D structures, capture fine-grained details, and generalize to novel objects using only category-level labels from seen categories. To address this, Chirpy3D introduces a multi-view diffusion model that decomposes training objects into anchor parts in an unsupervised manner, representing the latent space of both seen and unseen parts as continuous distributions. This allows smooth interpolation and flexible recombination of parts to generate entirely new objects with species-specific details. A self-supervised feature consistency loss further ensures structural and semantic coherence. The result is the first system capable of generating entirely novel 3D objects with species-specific fine-grained details through flexible part sampling and composition. Our experiments demonstrate that Chirpy3D surpasses existing methods in generating creative 3D objects with higher quality and fine-grained details. Code will be released at https://github.com/kamwoh/chirpy3d.