Blogs

3D Robotics

Review of the $50 Jevois computer vision camera/computer

Posted by Chris Anderson on May 7, 2017 at 3:30pm

3689715489?profile=original

Full review at our sister site, DIY Robocars, here, but this is how it starts:

I’m a huge fan of the OpenMV cam, which is a very neat $65 integrated camera and processor with sophisticated built-in computer vision libraries, a Micopython interpreter, and a very slick IDE (a car powered entirely by it came in 2nd in the Thunderhill DIY Robocars race). Now there is a competitor on the block, Jevois, which offers even more power and a lower cost. I’ve now spent a week with it and can report back on how it compares.

3689715631?profile=original

I tested it on an autonomous rover, to see how it performed in a stand-along embedded environment (as opposed to being connected via USB to a PC):

3689715715?profile=original

Read the rest here...

E-mail me when people leave their comments –
Follow

You need to be a member of diydrones to add comments!

Join diydrones

Comments

  • Greg Dronsky May 8, 2017 at 1:03pm

    Very interesting educational kit for $50. This is a good video that shows a few advantages over a RPi set:

    https://youtu.be/7cTfOckkGlE

  • Gary McCray May 7, 2017 at 5:41pm

    Good to know, I had a suspicion that although Nvidias deep learning big stuff is getting a lot of attention, they TK / TX boards still weren't getting much help on the open source level anyway.

    So as it stands, looks like the RPI-3 might be the way to go.

    At least sounds like it has the easiest entry and learning curve.

    And the easiest to append your own stuff too.

    Finishing up some major tractor / backhoe work in the back yard (clearing 40 year coastal rain forest brush and trees), so will be getting going with my 2 Lightware LM20's soon.

    Thanks for the info.

    Best,

    Gary

  • 3D Robotics
    Chris Anderson May 7, 2017 at 4:52pm

    Some guys in our group are using the TX2, and so far it's been nothing but distro pain for them. Nothing compiles as it should and even getting OpenCV running is a challenge. Not ready for prime time. 

  • Gary McCray May 7, 2017 at 4:03pm

    Thank you for putting together this review Chris it is very informative.

    One of the problems we have is that a lot of the times the nifty new hardware doesn't tell us much about implementation.

    This very much gives you an idea of what works and what doesn't and how easy to use it is.

    How about doing something like this for a TX2 and the somewhat previous Blog with FlytOS providing object bounding and recognition.

    Seriously are any of the guys in the Robocars group working on that?

    Best Regards,

    Gary

This reply was deleted.