From Depth Scanner to Corner Detection Technology
The LiDAR sensor on recent iPhones and iPad Pro models was designed mainly for depth mapping and faster autofocus. Researchers at the MIT Media Lab have just shown it can do something far more surprising: it can help your device effectively see around corners. This class of imaging, known as non-line-of-sight (NLOS) imaging, has typically required bulky, expensive lab lasers with little connection to everyday gadgets. MIT’s work proves that consumer-grade LiDAR, like the iPhone LiDAR sensor, is already capable of detecting and tracking objects completely outside the camera’s field of view. Instead of just measuring how far away visible surfaces are, the sensor’s reflected light is processed to infer the presence and motion of hidden objects. That makes your phone’s LiDAR less of a simple depth aid and more of a window into advanced computer vision and future smartphone AR capabilities.
How an iPhone LiDAR Sensor Can See Around Corners
MIT’s system doesn’t produce clear photographs of whatever is hiding around a corner. Instead, it builds up increasingly detailed clues. As you move your device, the LiDAR sends out pulses of light that bounce off visible walls and then scatter toward hidden objects before returning. The team calls their method an aperture sampling model: the sensor collects many noisy, partial LiDAR readings from slightly different viewpoints as the camera moves. Over time, software stitches these samples together to estimate an object’s presence, approximate shape, position, and motion. The result is a kind of optical echolocation—using light rather than sound to make sense of what the camera cannot see directly. This motion-aware processing is what turns ordinary, low-power hardware into practical corner detection technology that could eventually run on phones and tablets we already carry.
What the System Can Detect Right Now
In its current research form, the system is already capable of several impressive tasks. First, it can track a single hidden object as it moves, updating its inferred position in real time. Second, it can reconstruct that object’s rough shape, even though it remains outside the camera’s direct line of sight. Third, it can monitor multiple objects at once, separating their movements and positions in the hidden scene. Finally, and perhaps most importantly for robots and autonomous devices, it can perform camera self-localization using hidden landmarks. That means a robot could orient itself and navigate by referencing objects that are not directly visible. While this is still a research prototype rather than a polished consumer app, it strongly suggests that seeing around corners will eventually be a standard part of smartphone AR capabilities and robot perception systems.
Why Cheap, Accessible Hardware Is a Big Deal
Earlier NLOS imaging experiments relied on high-powered, specialized lasers that were too expensive and fragile for everyday use. MIT’s breakthrough is that all of this corner detection technology works with consumer-grade LiDAR hardware, similar to what is already embedded in modern smartphones. According to the researchers, the sensor setup they used can be put together for under USD 50 (approx. RM230), showing that the barrier to entry is now remarkably low. Even though phone makers do not currently expose the raw LiDAR data needed to replicate the demo directly on an iPhone, the underlying capability is there. As manufacturers open up deeper sensor access, the same type of processing could run on existing devices, turning today’s depth sensors into powerful tools for hidden-scene understanding, navigation, and advanced AR interactions.
Future Apps: AR, Safety, and Smarter Devices
Because the MIT team has released its code publicly, developers and hardware makers can immediately begin experimenting with corner-aware prototypes using compatible LiDAR setups. Looking ahead, this could transform smartphone AR capabilities: games and experiences might respond to people or objects before they actually enter the camera frame, or place virtual content based on hidden real-world geometry. Safety and navigation are equally promising. Delivery robots, wheelchairs, and small home robots could avoid collisions using objects they cannot see directly. Cars and drones might gain extra reaction time by sensing motion around blind corners. Even simple smartphone apps could provide alerts when something is approaching an intersection ahead. In short, turning the iPhone LiDAR sensor into a tool that can see around corners marks a major leap from simple depth sensing toward predictive, environment-aware devices.