Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Pu, Shu, Zeng, Boya, Zhou, Kaichen, Wang, Mengyu, Liu, Zhuang
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2512.23628
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914224061546496
author Pu, Shu
Zeng, Boya
Zhou, Kaichen
Wang, Mengyu
Liu, Zhuang
author_facet Pu, Shu
Zeng, Boya
Zhou, Kaichen
Wang, Mengyu
Liu, Zhuang
contents Generative models are increasingly used in 3D vision to synthesize novel shapes, yet it remains unclear whether their generation relies on memorizing training shapes. Understanding their memorization could help prevent training data leakage and improve the diversity of generated results. In this paper, we design an evaluation framework to quantify memorization in 3D generative models and study the influence of different data and modeling designs on memorization. We first apply our framework to quantify memorization in existing methods. Next, through controlled experiments with a latent vector-set (Vecset) diffusion model, we find that, on the data side, memorization depends on data modality, and increases with data diversity and finer-grained conditioning; on the modeling side, it peaks at a moderate guidance scale and can be mitigated by longer Vecsets and simple rotation augmentation. Together, our framework and analysis provide an empirical understanding of memorization in 3D generative models and suggest simple yet effective strategies to reduce it without degrading generation quality. Our code is available at https://github.com/zlab-princeton/3d_mem.
format Preprint
id arxiv_https___arxiv_org_abs_2512_23628
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Memorization in 3D Shape Generation: An Empirical Study
Pu, Shu
Zeng, Boya
Zhou, Kaichen
Wang, Mengyu
Liu, Zhuang
Computer Vision and Pattern Recognition
Machine Learning
Generative models are increasingly used in 3D vision to synthesize novel shapes, yet it remains unclear whether their generation relies on memorizing training shapes. Understanding their memorization could help prevent training data leakage and improve the diversity of generated results. In this paper, we design an evaluation framework to quantify memorization in 3D generative models and study the influence of different data and modeling designs on memorization. We first apply our framework to quantify memorization in existing methods. Next, through controlled experiments with a latent vector-set (Vecset) diffusion model, we find that, on the data side, memorization depends on data modality, and increases with data diversity and finer-grained conditioning; on the modeling side, it peaks at a moderate guidance scale and can be mitigated by longer Vecsets and simple rotation augmentation. Together, our framework and analysis provide an empirical understanding of memorization in 3D generative models and suggest simple yet effective strategies to reduce it without degrading generation quality. Our code is available at https://github.com/zlab-princeton/3d_mem.
title Memorization in 3D Shape Generation: An Empirical Study
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2512.23628