Ads
Pokémon Go, the popular augmented-reality game released by Niantic in 2016, has been the subject of controversy since its inception. Many have speculated that the game was created as a means to collect data from its users in order to develop an AI-based navigation system that could potentially rival Google Maps. This theory has gained significant traction in recent years as Niantic has developed an extensive database of real-world locations through the game.
Niantic, the parent company of Pokémon Go, has been working on a Visual Positioning System (VPS) that utilizes images of real-world locations collected through the game. By strategically placing PokéStops and Gyms in various physical locations, the company has been able to acquire images and scans of places that even Google Maps does not have street view images of. This has allowed Niantic to develop a more detailed model of streets, footpaths, parks, and businesses by using images captured by individuals walking around with their phones.
According to a blog post by Niantic, the company has developed in-depth scans of over 10 million locations worldwide, with users contributing approximately one million new individual scans each week. These images are being used to train over 50 million neural networks, which compress thousands of mapped images into digital recreations of real-world physical spaces. The ultimate goal of these scans is to enhance autonomous systems, navigation, and augmented reality products.
The technology behind Pokémon Go has been instrumental in developing Niantic’s VPS over the past five years. By using a single image from a phone to determine its position and orientation, the system generates a 3D map of the surrounding area. This has allowed Niantic to create a vast database of images that can be used to create more accurate and detailed representations of real-world locations.
For example, the blog post explains that if you are standing behind a church, the local model closest to you may only have observed the front entrance of that church and therefore may not be able to provide you with an accurate location. However, on a global scale, the company has observed that local models have captured thousands of churches around the world, allowing for a more comprehensive understanding of these types of structures.
The data collected through Pokémon Go has also been used to determine optimal walking paths between locations and to learn the routines of its users. By analyzing this data, Niantic has been able to make significant advancements in the development of its AI-based navigation system. This system has the potential to revolutionize the way we navigate the world around us, providing more accurate and detailed information than ever before.
In conclusion, Pokémon Go may have started as a simple augmented-reality game, but it has evolved into a powerful tool for collecting data and developing advanced technologies. Niantic’s VPS is just one example of how this data can be used to enhance various products and services. As the company continues to expand its database of real-world locations, we can expect to see even more innovative applications of this technology in the future.
