BeagleBone Deep Learning Video Demo

BeagleBoard often gets eclipsed by Raspberry Pi. Where the Pi focuses on ease-of-use, the BeagleBone generally has more power for hardcore applications. With machine learning AI all the rage now, BeagleBoard now has the BeagleBone AI, a board with specific features aimed at machine learning. A recent video (see below) shows a demo of using TIDL (Texas Instruments Deep Learning Library). The video includes an example of streaming video to a browser and using predefined learning models to identify things picked up by a web camera.

The CPU onboard is the TI Sitara AM5729. That’s a dual Arm Cortex A15 running at 1.5 GHz. There are also two C66x floating-point DSP processors and two dual ARM Cortex M4 coprocessors. Still need more? You get four embedded vision engines, two dual-core real-time units, a 2D graphics accelerator, a 3D graphics accelerator, and a subsystem for encoding and decoding video and cryptography.

There’s quite a bit of memory, too. In addition to off-chip memory, there’s 2.5 MB of memory on the device.

The demo doesn’t require any new software on the board. You don’t even need a monitor for the board since the video streams to an external web browser. Pretty cool.

We covered this device back when it was first announced. One thing we really like about the Beagles is the prehiperal real-time units.

10 thoughts on “BeagleBone Deep Learning Video Demo

  1. driving me nuts with the whole “have to click to see what the hack is” style of summaries lately, instead of spending your first paragraph telling me that a beagle bone and a pi are single board computers, why not tell me what people are doing with them… and if clairfication is needed do so later.

    An Arduino Powered Sexbot and Beer Replicator
    The lovely Arduino is a microcontroller board typically powered by an 8 bit AVR and has a USB interface similar to the Raspberry PI, using software on one’s own pc you can create a wide variety of projects. That’s when [Rock Johnson] came up with the idea of (Read More)

  2. I have to believe that the lower cost of entry is another factor, and what seems to be a larger web of support,m ay be why the raspberry Pi is popular. If a Chevy fulfill your expectations and needs, why buy a Caddy?

    1. What self respecting pimp DOESN’T ride in a caddy?

      Me I’m waiting for an arm or preferably RISCV that is at least 3GHz 8+ cores to replace my regular desktop for development work and be able to compile code in an acceptable amount of time without having to use x86_64 and cross compile. Oh, and have pcie so I can use a real gpu.

  3. “BeagleBoard often gets eclipsed by Raspberry Pi. Where the Pi focuses on ease-of-use, the BeagleBone generally has more power for hardcore applications.”

    A RPi 4 and a Teensy 4 may well be cheaper or/and more performant depending on your needs (quad A72 and a M7).

    That said, this is a cool board, and even if the above is true, it’s one board (less clutter) and it’s probably easier to get up and running and a complete unit.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.