Transferring multiscale map styles using generative adversarial networks

@article{Kang2019TransferringMM,
  title={Transferring multiscale map styles using generative adversarial networks},
  author={Yuhao Kang and Song Gao and Robert E. Roth},
  journal={International Journal of Cartography},
  year={2019},
  volume={5},
  pages={115 - 141}
}
ABSTRACT The advancement of the Artificial Intelligence (AI) technologies makes it possible to learn stylistic design criteria from existing maps or other visual art and transfer these styles to make new digital maps. In this paper, we propose a novel framework using AI for map style transfer applicable across multiple map scales. Specifically, we identify and transfer the stylistic elements from a target group of visual examples, including Google Maps, OpenStreetMap, and artistic paintings, to… Expand
GENERATIVE ADVERSARIAL NETWORKS TO GENERALISE URBAN AREAS IN TOPOGRAPHIC MAPS
  • A. Courtial, G. Touya, X. Zhang
  • Computer Science
  • The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
  • 2021
TLDR
This article presents how a generative adversarial network (GAN) can be employed to produce a generalised map that combines several cartographic themes in the dense context of urban areas, and highlights the impact of data and representation choices on the quality of predicted images, and the challenge of learning geographic relationships. Expand
INFERRING THE SCALE AND CONTENT OF A MAP USING DEEP LEARNING
TLDR
This paper proposes a method to infer the scale and the content of the map from its image, using convolutional neural networks trained with a few hundred maps from French geography textbooks, and results show promising results. Expand
Constraint Based Evaluation of Generalized Images Generated by Deep Learning
TLDR
This article proposes an adaptation of the constraint-based evaluation to the images generated by deep learning, and six raster-based constraints are proposed for a mountain road generalisation use case. Expand
S2OMGAN: Shortcut from Remote Sensing Images to Online Maps
TLDR
This work proposes a semi-supervised structure-augmented online map GAN (S$^{2}$OMGAN) model to generate online maps directly from remote sensing images and proposes edge structural similarity index (ESSI) as a metric to evaluate the quality of topological consistency between generated online maps and ground truths. Expand
Towards Cartographic Knowledge Encoding with Deep Learning
Map generalization, the process of simplifying detailed maps to smaller scale maps with the major semantic and structural characteristics maintained, plays a key role in multiscale cartographicExpand
MasterplanGAN: Facilitating the smart rendering of urban master plans via generative adversarial networks
  • Xinyue Ye, Jiaxin Du, Yu Ye
  • Computer Science
  • Environment and Planning B: Urban Analytics and City Science
  • 2021
TLDR
The achievements of this study contribute to the development of automatic generation of previously subjective and experience-oriented processes, which can serve as a useful tool for urban designers and planners to save time in real projects. Expand
Generating Historical Maps from Online Maps
  • Zekun Li
  • Computer Science
  • SIGSPATIAL/GIS
  • 2019
TLDR
An automatic system to generate a large amount of data for the training of text detection systems for historical maps using a conditional GAN model, to generate realistic historical map images from existing geographic datasets to overcome the challenge of content mismatch. Expand
GANmapper: geographical content filling
TLDR
A new method to create spatial data using a generative adversarial network (GAN) that enables generating approximate maps of the urban form and it is generalisable to augment other types of geoinformation, enhancing the completeness and quality of spatial data infrastructure. Expand
Detection of Pictorial Map Objects with Convolutional Neural Networks
TLDR
In this work, realistically drawn objects are identified on digital maps by convolutional neural networks by training binary classifiers based on Xception and InceptionResNetV2 to separate maps and pictorial maps. Expand
Segmentation and sampling method for complex polyline generalization based on a generative adversarial network
TLDR
The requirements for sampled images to ensure the effective learning of complex polyline generalization are analysed are analysed and the results show clear trends in generalization. Expand
...
1
2
...

References

SHOWING 1-10 OF 72 REFERENCES
Generative Adversarial Text to Image Synthesis
TLDR
A novel deep architecture and GAN formulation is developed to effectively bridge advances in text and image modeling, translating visual concepts from characters to pixels. Expand
GeoGAN: A Conditional GAN with Reconstruction and Style Loss to Generate Standard Layer of Maps from Satellite Images
TLDR
This model translates the satellite image to the corresponding standard layer map image using three main model architectures: a conditional Generative Adversarial Network (GAN), a conditional GAN which compresses the images down to a learned embedding, and a generator which is trained as a normalizing flow (RealNVP) model. Expand
Image-to-Image Translation with Conditional Adversarial Networks
TLDR
Conditional adversarial networks are investigated as a general-purpose solution to image-to-image translation problems and it is demonstrated that this approach is effective at synthesizing photos from label maps, reconstructing objects from edge maps, and colorizing images, among other tasks. Expand
Image Style Transfer Using Convolutional Neural Networks
TLDR
A Neural Algorithm of Artistic Style is introduced that can separate and recombine the image content and style of natural images and provide new insights into the deep image representations learned by Convolutional Neural Networks and demonstrate their potential for high level image synthesis and manipulation. Expand
Spatial interpolation using conditional generative adversarial neural networks
TLDR
A novel deep learning architecture named conditional encoder-decoder generative adversarial neural networks (CEDGANs) for spatial interpolation is designed, therein combining the encoder -decoder structure with adversarial learning to capture deep representations of sampled spatial data and their interactions with local structural patterns. Expand
Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks
TLDR
This work presents an approach for learning to translate an image from a source domain X to a target domain Y in the absence of paired examples, and introduces a cycle consistency loss to push F(G(X)) ≈ X (and vice versa). Expand
Satellite Image Spoofing: Creating Remote Sensing Dataset with Generative Adversarial Networks (Short Paper)
TLDR
The empirical results show that GANs can 'remember' a certain 'sense of place' and further apply that 'sense' to another place, and raise both public and GIScientists' awareness in the potential occurrence of fake satellite images, and its impacts on various geospatial applications, such as environmental monitoring, urban planning, and land use development. Expand
Conditional Generative Adversarial Nets
TLDR
The conditional version of generative adversarial nets is introduced, which can be constructed by simply feeding the data, y, to the generator and discriminator, and it is shown that this model can generate MNIST digits conditioned on class labels. Expand
Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
The architecture introduced in this paper learns a mapping function G : X 7→ Y using an adversarial loss such thatG(X) cannot be distinguished from Y , whereX and Y are images belonging to twoExpand
What is it like down there?: generating dense ground-level views and image features from overhead imagery using conditional generative adversarial networks
TLDR
This work is the first work to use cGANs to generate ground-level views given overhead imagery in order to explore the benefits of the learned representations, and shows that dense feature maps generated using the framework are more effective for land-cover classification than approaches which spatially interpolate features extracted from sparse ground- level images. Expand
...
1
2
3
4
5
...