Traditional map-making relies heavily on Geographic Information Systems (GIS), requiring domain expertise and being time-consuming, especially for repetitive tasks. Recent advances in generative AI (GenAI), particularly image diffusion models, offer new opportunities for automating and democratizing the map-making process. However, these models struggle with accurate map creation due to limited control over spatial composition and semantic layout. To address this, we integrate vector data to guide map generation in different styles, specified by the textual prompts. Our model is the first to generate accurate maps in controlled styles, and we have integrated it into a web application to improve its usability and accessibility. We conducted a user study with professional cartographers to assess the fidelity of generated maps, the usability of the web application, and the implications of ever-emerging GenAI in map-making. The findings have suggested the potential of our developed application and, more generally, the GenAI models in helping both non-expert users and professionals in creating maps more efficiently. We have also outlined further technical improvements and emphasized the new role of cartographers to advance the paradigm of AI-assisted map-making.

Most image-to-image translation models postulate that a unique correspondence exists between the semantic classes of the source and target domains. However, this assumption does not always hold in real-world scenarios due to divergent distributions, different class sets, and asymmet- rical information representation. As conventional GANs attempt to generate images that match the distribution of the target domain, they may hallucinate spurious instances of classes absent from the source domain, thereby dimin- ishing the usefulness and reliability of translated images. CycleGAN-based methods are also known to hide the mis- matched information in the generated images to bypass cy- cle consistency objectives, a process known as steganogra- phy. In response to the challenge of non-bijective image translation, we introduce StegoGAN, a novel model that leverages steganography to prevent spurious features in generated images. Our approach enhances the semantic consistency of the translated images without requiring ad- ditional postprocessing or supervision. Our experimental evaluations demonstrate that StegoGAN outperforms existing GAN-based models across various non-bijective image- to-image translation tasks, both qualitatively and quantita- tively. Our code and pretrained models are accessible at https://github.com/sian-wusidi/StegoGAN.

Numerous historical research endeavors necessitate synthesizing information derived from multiple historical maps. Linking geographic entities across diverse maps has posed a persistent challenge for researchers. The conventional approach involves a two-step pipeline, initially detecting entities within individual maps and subsequently associating entities from different maps through a linking post-processing step. This paper introduces an innovative solution by conceptualizing historical maps as a 3D spatiotemporal volume. It advocates for simultaneous segmentation and association by employing video instance segmentation. The experimental findings demonstrate the superiority of the proposed method over the traditional approach, achieving a notable 0.2 F1-score improvement. Furthermore, its end-toend nature significantly streamlines the workflow for geographic entity alignment, thereby substantially enhancing the level of automation.

Historical maps provide useful spatio-temporal information on the Earth's surface before modern earth observation techniques came into being. To extract information from maps, neural networks, which gain wide popularity in recent years, have replaced hand-crafted map processing methods and tedious manual labor. However, aleatoric uncertainty, known as data-dependent uncertainty, inherent in the drawing/scanning/fading defects of the original map sheets and inadequate contexts when cropping maps into small tiles considering the memory limits of the training process, challenges the model to make correct predictions. As aleatoric uncertainty cannot be reduced even with more training data collected, we argue that complementary spatio-temporal contexts can be helpful. To achieve this, we propose a U-Net-based network that fuses spatio-temporal features with cross-attention transformers (U-SpaTem), aggregating information at a larger spatial range as well as through a temporal sequence of images. Our model achieves a better performance than other state-or-art models that use either temporal or spatial contexts. Compared with pure vision transformers, our model is more lightweight and effective. To the best of our knowledge, leveraging both spatial and temporal contexts have been rarely explored before in the segmentation task. Even though our application is on segmenting historical maps, we believe that the method can be transferred into other fields with similar problems like temporal sequences of satellite images. Our code is freely accessible at https://github.com/chenyizi086/wu.2023.sigspatial.git.

Shaded relief is an effective method for visualising terrain on topographic maps, especially when the direction of illumination is adapted locally to emphasise individual terrain features. However, digital shading algorithms are unable to fully match the expressiveness of hand-crafted masterpieces, which are created through a laborious process by highly specialised cartographers. We replicate hand-drawn relief shading using U-Net neural networks. The deep neural networks are trained with manual shaded relief images of the Swiss topographic map series and terrain models of the same area. The networks generate shaded relief that closely resemble hand-drawn shaded relief art. The networks learn essential design principles from manual relief shading such as removing unnecessary terrain details, locally adjusting the illumination direction to accentuate individual terrain features, and varying brightness to emphasise larger landforms. Neural network shadings are generated from digital elevation models in a few seconds, and a study with 18 relief shading experts found that they are of high quality.

The design of new map projections has up until now required mathematical and cartographic expertise that has limited this activity to a small group of specialists. This article introduces the background mathematics for a software-based method that enables cartographers to easily design new small-scale world map projections. The software is usable even by those without mathematical expertise. A new projection is designed interactively in an iterative process that allows the designer to graphically and numerically assess the graticule, the representation of the continents, and the distortion properties of the new projection. The method has been implemented in Flex Projector, a free and open-source application enabling users to quickly create new map projections and modify existing projections. We also introduce new tools that help evaluate the distortion properties of projections, namely a configurable acceptance index to assess areal and angular distortion, a derived acceptance visualization, and interactive profiles through the distortion space of a projection. To illustrate the proposed method, a new projection, the Cropped Ginzburg VIII projection, is presented.