MetroGAN: Simulating Urban Morphology with Generative Adversarial Network

模拟具有位置属性的城市形态是城市科学中一项具有挑战性的任务。最近的研究表明，生成对抗网络(GANs)有可能揭示这一任务。然而，现有的基于gan的模型受限于城市数据的稀疏性和模型训练的不稳定性，阻碍了其应用。本文提出一种融合地理知识的GAN框架，即大都市GAN (MetroGAN)，用于城市形态模拟。纳入了一个渐进式增长结构来学习分层特征，并设计了一个地理损失来施加水域的限制。此外，本文还提出了城市系统复杂结构的综合评价框架。结果表明，MetroGAN在所有指标上都比目前最先进的城市模拟方法高出20%以上。令人鼓舞的是，仅使用自然地理特征，MetroGAN仍然可以生成城市的形状。实验结果表明，MetroGAN解决了以往城市模拟GANs的不稳定性问题，并且对各种城市属性具有通用性

MetroGAN: Simulating Urban Morphology with Generative Adversarial Network相关推荐

GAN（Generative Adversarial Network，GAN）模型应用
GAN(Generative Adversarial Network,GAN)模型人脸应用目录 GAN(Generative Adversarial Network,GAN)模型人脸应用人脸重构: ...
GAN（Generative Adversarial Network，GAN）模型之：SeqGAN、IRGAN、StackGAN、BigBiGAN、GAIN模型
GAN(Generative Adversarial Network,GAN)模型之:SeqGAN.IRGAN.StackGAN.BigBiGAN.GAIN模型目录 GAN(Generative A ...
GAN（Generative Adversarial Network，GAN）模型之：EBGAN、PGGAN、CGAN、ACGAN模型
GAN(Generative Adversarial Network,GAN)模型之:EBGAN.PGGAN.CGAN.ACGAN模型 EBGAN EBGAN的思想是,将D网络看做一个能量方程.当G网 ...
GAN（Generative Adversarial Network，GAN）模型之：InfoGAN、WGAN、DCGAN模型
GAN(Generative Adversarial Network,GAN)模型之:InfoGAN.WGAN.DCGAN模型 InfoGAN 为了解决定制化生成的问题,提出了InfoGAN模型.In ...
生成对抗网络（Generative Adversarial Network，GAN）
生成对抗网络(Generative Adversarial Network,GAN) Generative:生成式模型 Adversarial:采取对抗的策略 Networks:网络(不一定是深度学习 ...
DivCo: Diverse Conditional Image Synthesis via Contrastive Generative Adversarial Network
模式崩溃是cGAN较为重要的问题,以往工作大多数都致力于关注latent code和生成图像之间的关系而没有关注通过不同latent code生成的图像之间的关系.MSGAN(mode seeking ...
Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network
Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network Ledig C, Theis ...
PDGAN: A Novel Poisoning Defense Method in Federated Learning Using Generative Adversarial Network笔记
前言论文 "PDGAN: A Novel Poisoning Defense Method in Federated Learning Using Generative Adversari ...
论文阅读：Single Image Dehazing via Conditional Generative Adversarial Network
Single Image Dehazing via Conditional Generative Adversarial Network Runde Li∗ Jinshan Pan∗ Zechao L ...

MetroGAN: Simulating Urban Morphology with Generative Adversarial Network

MetroGAN: Simulating Urban Morphology with Generative Adversarial Network相关推荐

最新文章

热门文章