The unaided human eye will never be able to see around corners, but anything is possible with sufficiently sophisticated imaging technology. So-called line-of-sight (or NLOS) technology is an increasingly common area of ​​study in the age of self-driving cars, which would benefit tremendously from being able to see what’s going on around the bend. Now, a team at the Stanford Computational Imaging Lab has taken the idea a step further by spying on objects inside a locked room. All they need is a laser and a keyhole.

NLOS systems do not yet have real-world applications, but they could potentially help first responders locate people after a disaster and improve medical imaging. The idea has been validated many times, but current methods are too slow to be useful. In the past, NLOS experiments have relied on flat, reflective surfaces that can bind light around a barrier. The light (usually laser pulses) reflects off the target and returns to the experimenters. With some math we can get a rough idea of ​​what’s on the other side of the fence without looking.

The Stanford study does nothing on the issue of speed, but it works under a wider variety of conditions. The researchers simply shone a laser through the keyhole, projecting the dot onto the back wall. The photons from this beam are scattered, bouncing around the room and all objects inside. Over time, a small number of these photons will eventually come back through the keyhole, but that alone is not enough to say what these photons “saw” inside.

The researchers found that a moving object alters the laser pulse data in such a way that it is some recoverable data. Using an AI algorithm, it is possible to extract a passable representation of what is inside the room.

These “estimates” of the hidden object shapes will not give you a crystal clear picture. As you can see above, people end up looking a bit like birds. The letters are almost readable, however. The important thing is that these estimates were generated without ever looking directly into the subject. Under the right conditions with a refined algorithm, it might be possible to tell if someone is walking behind a closed door. Maybe it’s time to put duct tape on your keyholes.

Now read:

The unaided human eye will never be able to see around corners, but anything is possible with sufficiently sophisticated imaging technology. So-called line-of-sight (or NLOS) technology is an increasingly common area of ​​study in the age of self-driving cars, which would benefit tremendously from being able to see what’s going on around the bend. Now, a team at the Stanford Computational Imaging Lab has taken the idea a step further by spying on objects inside a locked room. All they need is a laser and a keyhole.

NLOS systems do not yet have real-world applications, but they could potentially help first responders locate people after a disaster and improve medical imaging. The idea has been validated many times, but current methods are too slow to be useful. In the past, NLOS experiments have relied on flat, reflective surfaces that can bind light around a barrier. The light (usually laser pulses) reflects off the target and returns to the experimenters. With some math we can get a rough idea of ​​what’s on the other side of the fence without looking.

The Stanford study does nothing on the issue of speed, but it works under a wider variety of conditions. The researchers simply shone a laser through the keyhole, projecting the dot onto the back wall. The photons from this beam are scattered, bouncing around the room and all objects inside. Over time, a small number of these photons will eventually come back through the keyhole, but that alone is not enough to say what these photons “saw” inside.

The researchers found that a moving object alters the laser pulse data in such a way that it is some recoverable data. Using an AI algorithm, it is possible to extract a passable representation of what is inside the room.

These “estimates” of the hidden object shapes will not give you a crystal clear picture. As you can see above, people end up looking a bit like birds. The letters are almost readable, however. The important thing is that these estimates were generated without ever looking directly into the subject. Under the right conditions with a refined algorithm, it might be possible to tell if someone is walking behind a closed door. Maybe it’s time to put duct tape on your keyholes.

Now read: