US researchers use reflection for imaging that could be applied to self-driving cars

When a car is driving down a narrow city street, the glossy paint of a parked car or the reflection in a rearview mirror can help the driver see something otherwise hidden from view, such as a child behind a parked car. playing on the sidewalk.

New computer vision technology uses reflection to turn shiny objects into "cameras" (Image source: MIT)

According to foreign media reports, based on the above situation, researchers from the Massachusetts Institute of Technology (MIT) and Rice University in the United States created a computer vision technology that uses reflection imaging. Their approach uses reflections to turn shiny objects into "cameras," allowing users to see the world as if through the "lens" of everyday objects like ceramic coffee mugs or metal paperweights.

Using images of an object taken from different angles, the technology turns the object's surface into a virtual sensor that can capture reflections. The artificial intelligence (AI) system maps such reflections in a way that allows it to estimate the depth of the scene and capture new views that can only be seen from the perspective of the object. In this way, people can see around corners or beyond obstacles.

This approach is particularly useful for self-driving cars. For example, it could allow self-driving cars to use reflections from passing objects, such as streetlights or buildings, to see what's going on around a parked truck.