Geospatial data come from various sources, such as satellites, aircraft, and LiDAR. The variability of the source is not limited to the types of data acquisition techniques, as we have maps from different time periods. To incorporate these data for a coherent analysis, it is essential to first align different "styles" of geospatial data to its matching images that point to the same location on the surface of the Earth. In this paper, we approach the image registration as a two-step process of (1) extracting geospatial contents invariant to visual (and any other non-content-related) information, and (2) matching the data based on such (purely) geospatial contents. We hypothesize that a combination of $\beta$-VAE-like architecture [2] and adversarial training will achieve both the disentanglement of the geographic information and artistic styles and generation of new map tiles by composing the encoded geographic information with any artistic style.

Aerial imagery and its direct application to visual localization is an essential problem for many Robotics and Computer Vision tasks. While Global Navigation Satellite Systems (GNSS) are the standard default solution for solving the aerial localization problem, it is subject to a number of limitations, such as, signal instability or solution unreliability that make this option not so desirable. Consequently, visual geolocalization is emerging as a viable alternative. However, adapting Visual Place Recognition (VPR) task to aerial imagery presents significant challenges, including weather variations and repetitive patterns. Current VPR reviews largely neglect the specific context of aerial data. This paper introduces a methodology tailored for evaluating VPR techniques specifically in the domain of aerial imagery, providing a comprehensive assessment of various methods and their performance. However, we not only compare various VPR methods, but also demonstrate the importance of selecting appropriate zoom and overlap levels when constructing map tiles to achieve maximum efficiency of VPR algorithms in the case of aerial imagery. The code is available on our GitHub repository -- https://github.com/prime-slam/aero-vloc.

With global climate change, devastating hurricanes are occurring with higher frequency. After a hurricane, roads are often flooded or washed out, making them treacherous for motorists. According to The Weather Channel, almost two of every three U.S. flash flood deaths from 1995–2010, excluding fatalities from Hurricane Katrina, occurred in vehicles. During my Insight A.I. Fellowship, I designed a system that detects flooded roads and created an interactive map app. Using state of the art computer vision deep learning methods, the system automatically annotates flooded, washed out, or otherwise severely damaged roads from satellite imagery.

With global climate change, devastating hurricanes are occurring with higher frequency. After a hurricane, roads are often flooded or washed out, making them treacherous for motorists. According to The Weather Channel, almost two of every three U.S. flash flood deaths from 1995–2010, excluding fatalities from Hurricane Katrina, occurred in vehicles. During my Insight A.I. Fellowship, I designed a system that detects flooded roads and created an interactive map app. Using state of the art computer vision deep learning methods, the system automatically annotates flooded, washed out, or otherwise severely damaged roads from satellite imagery.

"With cities, it is as with dreams: everything imaginable can be dreamed, but even the most unexpected dream is a rebus that conceals a desire or, its reverse, a fear. Cities, like dreams, are made of desires and fears, even if the thread of their discourse is secret, their rules are absurd, their perspectives deceitful, and everything conceals something else." A project made during "Machine Learning for Artists workshop" with Gene Kogan @Opendotlab In this project, we trained a neural network to translate map tiles into generative satellite images. We trained individual models for several cities–Milan, Venice, and Los Angeles–allowing us to do city map style transfer (example above) by applying the aerial model of one city onto the map tiles of another. Also, we can create imaginary cities by hand-drawing sketches and feeding them to the generative model.

We were fascinated by the possibility of generating new and non-existent but realistic images using conditional adversarial neural networks that remembers a certain set of features from what it has seen in the past; the same process that we humans undergo when we dream. Taking inspiration from the given examples, we applied a pre-defined color scheme to geographic data (OpenStreetMap) using Mapbox Studio: roads, green spaces, buildings, water were styled with different colours (black, green, red, blue), so that the neural network (NN) could compare these to aerial images and learn the correspondence between them. We then used vvvv as a tool to collect both satellite imagery and the associated labeled map tiles. We trained a conditional generative adversarial network, using pix2pix to reconstruct the satellite imagery from its map tiles. The sample below on the right shows the reconstructed sattelite image of the given map tile, closely resembling to the original on the left.

Right now, Terrapattern only covers four American cities: Pittsburgh, Detroit, San Francisco, and New York City. Terrapattern is so computing-hungry that it is effectively a proof of concept right now, at least for a team of artists working with less than 35,000. Each metro region takes about 10 gigabytes of RAM--not storage, but active memory. That said, Terrapattern is relatively technically straightforward. It's constructed from a convolutional neural network and CoverTree, an algorithm that remembers some descriptions and allows the searches to happen quickly.