Skip to main content
  1. Home
  2. Emerging Tech
  3. Computing
  4. News

Kickstarter machine-vision camera gives your robot project eyes and a brain

Add as a preferred source on Google

You only need to pop open a website detailing the latest happenings in artificial intelligence (may we recommend Digital Trends?) to be aware that some exciting stuff is going on right now.

But while it’s all well and good to know that these breakthroughs are taking place in labs, and will one day trickle down to consumer products, what if you want to take advantage of some of them in your own DIY Arduino robot project — and without having to first earn a PhD in computer science?

Recommended Videos

Fortunately, a new Kickstarter project offers one possible answer. Created by Dr. Laurent Itti, professor of computer science, psychology and neuroscience at the University of Southern California, JeVois in a diminutive, open-source, quad-core smart machine vision camera designed for users to incorporate into their “maker” projects.

Coming preprogrammed with a bank of cutting-edge algorithms, the camera will not only record images, but will also let your robot project make sense of them. Results are streamed over USB to a host computer, or over a serial port to a micro-controller.

“A lot of progress has been made in machine learning and machine vision,” Dr. Itti told Digital Trends. “However, this is often still confined to the lab. What we want to do is to help people use some of these algorithms in their own projects by providing them with a platform which lets them experience them, without having to program them on their own. This is like having a smartphone but no programming experience. Instead, you just go to the App Store and download the apps that you need. You can still enjoy them, even if you don’t know how they work.”

Image used with permission by copyright holder

JeVois opens up an exciting range of possibilities and potential use-cases.

“Recognizing QR codes is one that everyone will understand,” he continued. “But there are other, more advanced things you can do as well. For example, you could get it to detect roads if you’re trying to build an autonomous car. We also have algorithms that can detect objects, algorithms that can help your robot work out which room it is in, and more.”

Some of the other applications not mentioned by Itti include human face detection, eye-tracking, object recognition in cluttered scenes, and a wealth of others. Many of the algorithms operate at 30, 60 or 120 frames per second, and if you have the necessary coding know-how, you can also program new algorithms yourself.

At present, JeVois is raising funds on Kickstarter, where you can pre-order a unit starting at $49 for the Barebones Kit. Shipping is set to start in February, with the majority of orders arriving in March.

Luke Dormehl
I'm a UK-based tech writer covering Cool Tech at Digital Trends. I've also written for Fast Company, Wired, the Guardian…
A new AI model wants self-driving cars to think before they swerve
Forget guessing games, this self-driving AI actually shows its work.
Self driving car

Self-driving cars have gotten pretty good at driving. The harder problem is teaching them to drive safely in situations nobody planned for. We have heard horror stories of self-driving cars, behaving erratically in emergency situations, often times delaying first responders from reaching the scene.

A team at Seoul National University, led by professor Jun Won Choi from the Department of Electrical and Computer Engineering, thinks they have cracked part of that puzzle with a new AI model called SafeDrive. The research was recently selected as a highlight paper at CVPR 2026, a distinction that goes only to roughly 3% of all submissions.

Read more
An AirPods-sized mini lab could help catch disease markers in minutes
Meet VPodDuo, an AirPods-sized molecular testing gadget
VPodDuo Mini Medical Device

Home testing could eventually become a lot more convenient. Researchs at the at the University of Illinois Urbana-Champaign has developed the VPodDuo, a pocket-sized fluorescence reader designed to measure sensitive molecular tests outside a conventional laboratory.

Its housing is roughly comparable in size and shape to an AirPods charging case, while Bluetooth connectivity sends results directly to a smartphone. Meaning, you might not have to wait in line anymore.

Read more
Your future AR glasses might finally stop making the real world look weird
Researchers have developed a new display technique that keeps virtual objects realistic without dimming the real world.
Samsung AR Glasses

One of the biggest promises of augmented reality is blending digital objects seamlessly into the real world. Ironically, today's AR glasses often make the real world harder to see. Researchers at the Shibaura Institute of Technology (SIT) in Japan, led by Assistant Professor Xiaodan Hu in collaboration with Dr. Yan Zhang and Professor Xubo Yang from Shanghai Jiao Tong University in China, and Professor Kiyoshi Kiyokawa from Japan's Nara Institute of Science and Technology, have developed a new AR display technique that aims to make virtual objects look realistic without reducing visibility of the real world. The study was published in IEEE Transactions on Visualization and Computer Graphics.

The biggest AR problem isn't virtual objects; it's everything around them

Read more