Thursday, March 05, 2020

AI in Image Sensor - Spectrum vs Nature

IEEE Spectrum publishes an article "Image Sensor Doubles as a Neural Net" about TU Wien, Austria, work on integration of AI engine onto an image sensor:

"Our image sensor does not consume any electrical power when it is operating," Lukas Mennel, an electrical engineer at TU Wien, says. "The sensed photons themselves provide the energy for the electric current."

Mennel notes the speed at which the system operates is only limited by the speed of the electrons in the circuits. In principle, this strategy could work on the order of picoseconds, or trillionths of a second, or three to four orders of magnitude faster than currently demonstrated, he says.

Nature too publishes comments to that same paper on neural network processor integrated onto an image sensor by Vienna University of Technology, Austria:

"Imaging under dim light would be difficult for the device described by the authors. A redesign would be needed to improve light absorption in the thin semiconductor and to increase the range of light intensities that can be detected. Furthermore, the reported design requires high voltages and consumes a lot of power; by comparison, the energy consumption per operation in a biological neural network is at the sub-femtojoule level (10e−15 to 10e−13 joules). It would also be useful to expand the response to ultraviolet and infrared light, to capture information unavailable in the visible spectrum."

No comments:

Post a Comment

All comments are moderated to avoid spam and personal attacks.