Google Maps Immersive View Could Turn Your Travels Into New Adventures

Artificial intelligence (AI) is making maps smarter and easier to use. 

Google Immersive View for Maps uses computer vision and AI to combine Street View and aerial photography into a 3D format. The software is now rolling out in five cities: London, Los Angeles, New York City, San Francisco, and Tokyo. Experts say the new map features are a sign of things to come. 

"AI increases the speed of decision-making by extracting the information needed to make the right decisions," Dustin Parkman, the vice president of transportation solutions at Bentley Systems, told Lifewire in an email interview. "For example, with AI-enabled mapping, consumer apps can scan road information and use algorithms to determine the optimal route to take, whether by foot or in a car, bike, bus, or train."

Google Immersive View, Powered By AI

Google claims its new Immersive View will help users get around by offering a more intuitive map. The company says it uses AI to fuse billions of Street View and aerial images. 

"With our new immersive view, you'll be able to experience what a neighborhood, landmark, restaurant or popular venue is like—and even feel like you're right there before you ever set foot inside," Miriam Daniel, the vice president of Google Maps Experiences wrote in a blog post on the company's website. "So whether you're traveling somewhere new or scoping out hidden local gems, immersive view will help you make the most informed decisions before you go."

Christopher Goranson, who teaches mapping and data visualization as a professor at Carnegie Mellon University, told Lifewire via email that AI can help find and categorize things quickly over large areas. For example, his students recently contributed to mapping efforts in response to the devastating earthquake that impacted Turkey and Syria. Using a system run by the Humanitarian OpenStreetMap Team, they used AI-generated data and satellite imagery to help identify and validate the existence of buildings and roads.  

"Their work and that of many other volunteers helps to improve the quality of the AI-detected data," Goranson said. "That information can provide a critical tool for emergency responders to identify buildings that were destroyed or to calculate estimates of where to target recovery efforts next, identify how many people were impacted, or find smaller, remote regions that were also impacted but that might not be getting as much attention." 

Google isn't the only company using AI maps. HERE Technologies is developing HD maps, machine-readable maps designed for automated vehicles rather than people. 

HERE uses AI models to automate the processing of 500 million kilometers of vehicle probe and sensor data every hour. The models extract map features such as 2D and 3D positioning of road signs. Other models are responsible for validating speed limits on the map. The models can also be trained in detecting and estimating traffic incidents and formulating ETA predictions.

"We also have trained algorithms to transform spatial data into map form," Giovanni Lanfranchi, Chief Technology and Product Officer at HERE Technologies, told Lifewire in an email interview. When we feed the HERE location platform with pictures, probe data, car camera feeds, lidar, and IoT data, it will rapidly transform and conflate that data into an actionable, navigable digital map."

The Future of AI Maps

AI expert Anantha Krishnan, the CEO of Sarva Labs, predicted in an email that in the future, AI will develop human-like intuition that will aid mapping.

"Say part of a four-lane road is occluded by trees," Krishnan said. "Humans can easily surmise that a couple lanes are hidden behind the trees. Today's machine-learning models most likely predict a two-lane road. New AI approaches try to mimic the natural behavior of humans to make better predictions."

As AI maps evolve, the technology may have its downsides. Goranson warned that  AI could also be used to exploit information in a way that negatively impacts people and communities.  

"More visibility in terms of how these tools can and are being used is crucial to ensuring that productive uses of AI are encouraged while incentives for bad actors are limited," he said.