Exploring in the Latent Space of Design: A Method of Plausible Building Facades Images Generation, Properties Control and Model Explanation Base on StyleGAN2

Meng, Shengyu

doi:10.1007/978-981-16-5983-6_6

Cited by 11 publications

(13 citation statements)

References 19 publications

(20 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…StyleGAN2 can achieve a certain degree of unsupervised high‐level attribute separation for generated images, 25 so that some style‐mixing operations can be performed. Meng 15 combined this feature and use principal component analysis (PCA) to visualize the process of generating mixed styles, which is beneficial to understand the correlation between images and their latent vectors. The vectors are randomly sampled in the high‐dimensional latent space Z and projected to the low‐dimensional latent space W by PCA to visualize the Z space and reflect the distribution of the features in Z .…”

Section: Interpretation Strategies For Gan Modelsmentioning

confidence: 99%

“…x is the scale parameter, which means high‐level attributes can be manipulated by moving the latent vector w in the W space along a certain principal axis to achieve manipulation of the attributes of the generated image. However, the generated image still has blurred details, and the attributes are not fully decoupled 15 …”

Section: Interpretation Strategies For Gan Modelsmentioning

confidence: 99%

“…However, the generated image still has blurred details, and the attributes are not fully decoupled. 15…”

Section: Pca Projectionmentioning

confidence: 99%

“…In addition, GAN has become one of the most promising methods for image generation, classification, detection, interpretation, text‐to‐image 12 and other applications due to its excellent ability to generate realistic images, which can generate photo‐realistic images and simulate facial aging 13,14 . Its variants also enable the generation of architectural perspective images and multimodal images 15,16 …”

Section: Introductionmentioning

confidence: 99%

“…13,14 Its variants also enable the generation of architectural perspective images and multimodal images. 15,16 However, without the dissection and interpretation of the generation process of GANs, we cannot understand why GANs can generate photo-realistic images and why the output images can be controlled by interpretable units. 17 We cannot control attributes of images as well, because other associated attributes may change if we control a single attribute.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Current status, application, and challenges of the interpretability of generative adversarial network models

Wang

Zhao

Huang

et al. 2022

Computational Intelligence

View full text Add to dashboard Cite

The generative adversarial network (GAN) is one of the most promising methods in the field of unsupervised learning. Model developers, users, and other interested people are highly concerned about the GAN mechanism where the generative model and the discriminative model learn from each other in a gameplay manner, which generates a causal relationship among output features, internal network structure, feature extraction process, and output results. Through the study of the interpretability of GANs, the validity, reliability, and robustness of the application of GANs can be verified, and the weaknesses of the GANs in specific applications can be diagnosed, which can provide support for designing better network structures. It can also improve security and reduce the decision‐making and prediction risks brought by GANs. In this article, the study of the interpretability of GANs is explored, and ways of the evaluation of the application effect of GAN interpretability techniques are analyzed. Besides, the effect of interpretable GANs in fields such as medical treatment and military is discussed, and current limitations and future challenges are demonstrated.

show abstract

Section: Interpretation Strategies For Gan Modelsmentioning

confidence: 99%

Section: Interpretation Strategies For Gan Modelsmentioning

confidence: 99%

“…However, the generated image still has blurred details, and the attributes are not fully decoupled. 15…”

Section: Pca Projectionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Current status, application, and challenges of the interpretability of generative adversarial network models

Wang

Zhao

Huang

et al. 2022

Computational Intelligence

View full text Add to dashboard Cite

show abstract

Local intelligence: time to learn from AI

Başarır,

Çiçek,

Koç

2024

Architectural Science Review

View full text Add to dashboard Cite

UDGAN: A new urban design inspiration approach driven by using generative adversarial networks

Gan,

Zhao,

Wang

et al. 2023

Journal of Computational Design and Engineering

View full text Add to dashboard Cite

The morphological design of urban space affects the quality of the environment. The traditional experience-based design approach was greatly improved by introducing computational design tools. However, the existing urban design tools are mostly developed on pre-set rules or given targets, which have few contribution to enhance creativity or generate inspiring schemes. Therefore, this paper proposes a new computational urban design approach named UDGAN, integrating generative adversarial networks (GANs) and multi-objective optimization algorithms. This model utilize urban design scheme plans over the past 20 years from a particular designer as training datasets. Four preference models were trained to autonomously generate stylized urban design schemes. Eight morphological parameters were used to analyze the model performance by comparing generated results with the ground truth. This GAN-based surrogate approach is combined with a morphological indicator alignment process using multi-objective optimization model to obtain better results. The result shows that the r2 predicted by the improved Pix2Pix model reaches 0.798, and the similarity of the generated results can be stably distributed between 0.7-0.8, so the design scheme of this preferred style can be effectively learned. At the same time, the pre-trained model greatly reduces the time consumption of the design scheme generation, taking 5 min approximately to complete a generation process. This approach quickly generated the design scheme with preferred features, supporting the designer with creativity and greatly saving the time of design creation, transforming computational design into an inspiration-driven process.

show abstract

Exploring in the Latent Space of Design: A Method of Plausible Building Facades Images Generation, Properties Control and Model Explanation Base on StyleGAN2

Cited by 11 publications

References 19 publications

Current status, application, and challenges of the interpretability of generative adversarial network models

Current status, application, and challenges of the interpretability of generative adversarial network models

Local intelligence: time to learn from AI

UDGAN: A new urban design inspiration approach driven by using generative adversarial networks

Contact Info

Product

Resources

About