Reconstructing 3D Objects With a Tiny Distance Sensor
1 min read
Summary
Researchers from the University of Washington have used a cheap, time-of-flight sensor and a neural network to create 3D scans of objects.
Although the sensor measures the time for light to bounce back to determine the distance to a surface or object, it is also possible to use the intensity of the return pulse over time, known as the transient histogram, to recreate 3D geometry, reported Hackaday.
This technique is still in its infancy and requires further refinement, but it could provide a viable, cheap alternative to 3D scanning, according to the researchers’ paper published on arXiv.org.
The system relies on compressed sensing followed by the use of a neural network to solve the reconstruction problem.
An acoustic equivalent to this visual technique has also recently been developed.
