GUNet: A Graph Convolutional Network United Diffusion Model for Stable and Diversity Pose Generation

Shuowen Liang, Sisi Li, Qingyun Wang, Cen Zhang, Kaiquan Zhu, Tian Yang
{"title":"GUNet: A Graph Convolutional Network United Diffusion Model for Stable and Diversity Pose Generation","authors":"Shuowen Liang, Sisi Li, Qingyun Wang, Cen Zhang, Kaiquan Zhu, Tian Yang","doi":"arxiv-2409.11689","DOIUrl":null,"url":null,"abstract":"Pose skeleton images are an important reference in pose-controllable image\ngeneration. In order to enrich the source of skeleton images, recent works have\ninvestigated the generation of pose skeletons based on natural language. These\nmethods are based on GANs. However, it remains challenging to perform diverse,\nstructurally correct and aesthetically pleasing human pose skeleton generation\nwith various textual inputs. To address this problem, we propose a framework\nwith GUNet as the main model, PoseDiffusion. It is the first generative\nframework based on a diffusion model and also contains a series of variants\nfine-tuned based on a stable diffusion model. PoseDiffusion demonstrates\nseveral desired properties that outperform existing methods. 1) Correct\nSkeletons. GUNet, a denoising model of PoseDiffusion, is designed to\nincorporate graphical convolutional neural networks. It is able to learn the\nspatial relationships of the human skeleton by introducing skeletal information\nduring the training process. 2) Diversity. We decouple the key points of the\nskeleton and characterise them separately, and use cross-attention to introduce\ntextual conditions. Experimental results show that PoseDiffusion outperforms\nexisting SoTA algorithms in terms of stability and diversity of text-driven\npose skeleton generation. Qualitative analyses further demonstrate its\nsuperiority for controllable generation in Stable Diffusion.","PeriodicalId":501130,"journal":{"name":"arXiv - CS - Computer Vision and Pattern Recognition","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Computer Vision and Pattern Recognition","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.11689","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Pose skeleton images are an important reference in pose-controllable image generation. In order to enrich the source of skeleton images, recent works have investigated the generation of pose skeletons based on natural language. These methods are based on GANs. However, it remains challenging to perform diverse, structurally correct and aesthetically pleasing human pose skeleton generation with various textual inputs. To address this problem, we propose a framework with GUNet as the main model, PoseDiffusion. It is the first generative framework based on a diffusion model and also contains a series of variants fine-tuned based on a stable diffusion model. PoseDiffusion demonstrates several desired properties that outperform existing methods. 1) Correct Skeletons. GUNet, a denoising model of PoseDiffusion, is designed to incorporate graphical convolutional neural networks. It is able to learn the spatial relationships of the human skeleton by introducing skeletal information during the training process. 2) Diversity. We decouple the key points of the skeleton and characterise them separately, and use cross-attention to introduce textual conditions. Experimental results show that PoseDiffusion outperforms existing SoTA algorithms in terms of stability and diversity of text-driven pose skeleton generation. Qualitative analyses further demonstrate its superiority for controllable generation in Stable Diffusion.
GUNet:用于生成稳定和多样化姿势的图卷积网络联合扩散模型
姿势骨架图像是姿势可控图像生成的重要参考。为了丰富骨架图像的来源,最近有研究基于自然语言生成姿势骨架。这些方法都基于 GAN。然而,利用各种文本输入生成多样化、结构正确且美观的人体姿态骨架仍然具有挑战性。为了解决这个问题,我们提出了一个以 GUNet 为主要模型的框架,即 PoseDiffusion。它是第一个基于扩散模型的生成框架,还包含一系列基于稳定扩散模型进行微调的变体。PoseDiffusion 展示了优于现有方法的多个理想特性。1) 正确的骨架。GUNet 是 PoseDiffusion 的去噪模型,其设计目的是结合图形卷积神经网络。它能够通过在训练过程中引入骨骼信息来学习人体骨骼的空间关系。2) 多样性。我们将骨骼的关键点分离开来,分别描述它们的特征,并使用交叉注意引入文本条件。实验结果表明,在文本驱动骨架生成的稳定性和多样性方面,PoseDiffusion 优于现有的 SoTA 算法。定性分析进一步证明了 PoseDiffusion 在稳定扩散的可控生成方面更胜一筹。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信