Repurposing 2D Diffusion Models for 3D Shape Completion

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: He, Yao, Kwon, Youngjoong, Xiang, Tiange, Cai, Wenxiao, Adeli, Ehsan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911322018414592
author He, Yao
Kwon, Youngjoong
Xiang, Tiange
Cai, Wenxiao
Adeli, Ehsan
author_facet He, Yao
Kwon, Youngjoong
Xiang, Tiange
Cai, Wenxiao
Adeli, Ehsan
contents We present a framework that adapts 2D diffusion models for 3D shape completion from incomplete point clouds. While text-to-image diffusion models have achieved remarkable success with abundant 2D data, 3D diffusion models lag due to the scarcity of high-quality 3D datasets and a persistent modality gap between 3D inputs and 2D latent spaces. To overcome these limitations, we introduce the Shape Atlas, a compact 2D representation of 3D geometry that (1) enables full utilization of the generative power of pretrained 2D diffusion models, and (2) aligns the modalities between the conditional input and output spaces, allowing more effective conditioning. This unified 2D formulation facilitates learning from limited 3D data and produces high-quality, detail-preserving shape completions. We validate the effectiveness of our results on the PCN and ShapeNet-55 datasets. Additionally, we show the downstream application of creating artist-created meshes from our completed point clouds, further demonstrating the practicality of our method.
format Preprint
id arxiv_https___arxiv_org_abs_2512_13991
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Repurposing 2D Diffusion Models for 3D Shape Completion
He, Yao
Kwon, Youngjoong
Xiang, Tiange
Cai, Wenxiao
Adeli, Ehsan
Computer Vision and Pattern Recognition
We present a framework that adapts 2D diffusion models for 3D shape completion from incomplete point clouds. While text-to-image diffusion models have achieved remarkable success with abundant 2D data, 3D diffusion models lag due to the scarcity of high-quality 3D datasets and a persistent modality gap between 3D inputs and 2D latent spaces. To overcome these limitations, we introduce the Shape Atlas, a compact 2D representation of 3D geometry that (1) enables full utilization of the generative power of pretrained 2D diffusion models, and (2) aligns the modalities between the conditional input and output spaces, allowing more effective conditioning. This unified 2D formulation facilitates learning from limited 3D data and produces high-quality, detail-preserving shape completions. We validate the effectiveness of our results on the PCN and ShapeNet-55 datasets. Additionally, we show the downstream application of creating artist-created meshes from our completed point clouds, further demonstrating the practicality of our method.
title Repurposing 2D Diffusion Models for 3D Shape Completion
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.13991