arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.06124cs.LGcs.CV
收藏

结构化生成:使用层次聚类引导扩散模型

Structured Generations: Using Hierarchical Clusters to guide Diffusion Models

ETH Zurich(苏黎世联邦理工学院)

Jorge da Silva Goncalves, Laura Manduchi, Moritz Vandenhirtz, Julia E. Vogt

AI总结:

本研究提出Diffuse-TreeVAE模型,将层次聚类融入DDPM框架,通过潜在树VAE结构采样与第二阶段DDPM优化,生成高质量且具簇代表性的图像,改进了基于VAE的生成方法。

摘要:

本文提出Diffuse-TreeVAE,一种将层次聚类融入去噪扩散概率模型(DDPMs)框架的深度生成模型。该方法通过从已学习的基于潜在树VAE的结构的根嵌入采样来生成新图像,随后沿层次路径传播,并利用第二阶段DDPM为每个数据簇优化并生成独特的高质量图像。最终模型不仅提升了图像清晰度,还确保生成样本能代表其对应簇,解决了以往基于VAE方法的局限性,推动了基于聚类的生成建模的发展。

英文摘要:

This paper introduces Diffuse-TreeVAE, a deep generative model that integrates hierarchical clustering into the framework of Denoising Diffusion Probabilistic Models (DDPMs). The proposed approach generates new images by sampling from a root embedding of a learned latent tree VAE-based structure, it then propagates through hierarchical paths, and utilizes a second-stage DDPM to refine and generate distinct, high-quality images for each data cluster. The result is a model that not only improves image clarity but also ensures that the generated samples are representative of their respective clusters, addressing the limitations of previous VAE-based methods and advancing the state of clustering-based generative modeling.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑