Artificial intelligence brings new life to once-thriving, now-extinct, often decimated cities and neighborhoods to the ground, at least roughly. This is how you can venture into the past.

Many older neighborhoods have already fallen victim to new real estate developments and highway construction. Thanks to a new project, we just have to put on a pair of virtual reality glasses, and we can actually see homes and streets that no longer exist in their old glory. And it’s all thanks to an innovative method that uses machine learning and maps to create 3D digital models of old neighbourhoods, he explains ZME Sciences.

Of course, the point of the project is not just to be able to venture into the past, but such digital models could be an important addition to research that was previously considered quite challenging. With the help of 3D models, specialists can make more accurate and transparent analyses, for example, regarding the economic effects of the destruction of a potentially historic part of a city.

See also  Economy: MÁVapp and vending machines have been suspended under the new tariff system

OSU specialists used Sanborn maps in their project. It was originally designed to help fire insurance companies assess risk exposures in thousands of cities across the United States in the 19th and 20th centuries. Researchers have developed machine learning tools that can extract information from the Sanborn maps, such as specific details about individual buildings, such as their exact location, number of floors, building materials, and their primary purpose, whether it is residential or commercial.




The new method was tested in two adjacent areas east of Columbus, Ohio. Its destruction was massive in the 1960s due to the construction of a highway. Not only will researchers visualize what was lost, but they also measure damaged capital and wealth-generating activities, as well as the resulting environmental problems.

They focused on a total of 13 Sanborn maps created in 1961, just before the highway was built. Machine learning techniques were used to extract data from these maps and produce digital models of neighborhoods. They discover that nearly 400 structures, such as homes and stores, were destroyed before the highway was built. The machine learning model has proven to be very accurate in reconfiguring information on maps. The researchers believe they will be able to create 3D models of the 12,000 cities and towns covered by Sandborn’s maps. Their way is Plus one Detailed in the journal.

If you want to know similar things other times, like it HVG Tech department’s Facebook page.

See also  Sasha Pieters gained more than 30 kilos in one year due to his undiagnosed hormonal disease




HVG


In addition to diverse, independent and factual information, our readers who join the Pártoló membership can also enjoy a number of benefits for their financial support.
Depending on your membership level, we offer, among others:

  • We send you an exclusive weekly digest of the interesting things in the world;
  • You can gain insight into the work of HVG, you can meet our authors;
  • You can take part in pre-premier screenings of the latest films, in various events;
  • You can buy HVG books and publications at a discount;
  • You can read hvg360 digital news magazine.