Kyocera triple lens AI depth camera enable recognition of thin and semi-transparent objects, wires

Kyocera has unveiled a triple lens AI depth camera capable of recognizing semi-transparent, thin, and fine line-shaped objects that are difficult to detect with the human eye and traditional stereo cameras.

The camera can accurately measure the distance to and size of objects which are between 0.3 and 1mm thick, and is expected to be useful in robots for manufacturing, medical applications, and Smart Agriculture.

Kyocera triple lens AI camera

Preliminary specifications:

  • Sensors
    • Left-center, center-right, and left-right cameras
    • Focus distance – About 10 cm
  • Proprietary AI combining multiple parallax data sets from the sensors
  • Ideal for
    • Thin, irregularly shaped linear objects, such as harnesses or ultra-fine wires as small as 0.3mm
    • Reflective objects like metal
    • Translucent objects like plastic
  • Dimensions – 40 x 30 x 28mm

It is an evolution of the company’s dual-lens AI depth camera that was already capable of  high-precision distance measurement with 100μm resolution at a 10cm range, but struggled a bit more for distance measurements and detection of ultra-thin objects or in environments where part of the object is obscured. Adding a third camera sensor coupled with an AI magic apparently solves that.

Ultra thin cable AI detection
Original image with 0.3mm cable (left), as seen by dual lens camera (center), and by triple lens camera (right)

The dual-lens depth sensor would be blind and completely miss the 0.3mm cable, but the triple camera can properly detect it and measure the distance to it (about 80mm). Another example shown below is with overlapping transparent plastic cubes that can better represented with a triple camera.

AI camera transparent cubes
Original image (left), dual lens camera (center), and triple lens camera (right)

The new triple lens AI depth camera will be useful for the inspection processes involving objects with highly repetitive patterns, such as electronic circuit boards or textiles, surgical robots whose needles and sutures are often hard to distinguish against their background and crop harvesting robots where fruits and leaves occlude each other.

Kyocera’s triple lens AI-based high-resolution depth sensor is not available, and will be demonstrated at CES 2026 in January along with other demos such as a demonstrate of underwater wireless optical communication at up to 5.2 Gbps. The dual lens camera model was demonstrated at CES 2025 earlier this year, but there’s no product page, and no availability or pricing information either.

Thanks to TLS for the tip.

Share this:

Support CNX Software! Donate via cryptocurrencies, become a Patron on Patreon, or purchase goods on Amazon or Aliexpress. We also use affiliate links in articles to earn commissions if you make a purchase after clicking on those links.

Radxa Orion O6 Armv9 mini-ITX motherboard
Subscribe
Notify of
guest
The comment form collects your name, email and content to allow us keep track of the comments placed on the website. Please read and accept our website Terms and Privacy Policy to post a comment.
0 Comments
oldest
newest
Boardcon MINI1126B-P AI vision system-on-module wit Rockchip RV1126B-P SoC