OAK-D Depth Sensing AI Camera Gets Smaller And Lighter

October 15, 2021 by Donald Papp 9 Comments

The OAK-D is an open-source, full-color depth sensing camera with embedded AI capabilities, and there is now a crowdfunding campaign for a newer, lighter version called the OAK-D Lite. The new model does everything the previous one could do, combining machine vision with stereo depth sensing and an ability to run highly complex image processing tasks all on-board, freeing the host from any of the overhead involved.

Animated face with small blue dots as 3D feature markers. — An example of real-time feature tracking, now in 3D thanks to integrated depth sensing.

The OAK-D Lite camera is actually several elements together in one package: a full-color 4K camera, two greyscale cameras for stereo depth sensing, and onboard AI machine vision processing with Intel’s Movidius Myriad X processor. Tying it all together is an open-source software platform called DepthAI that wraps the camera’s functions and capabilities together into a unified whole.

The goal is to give embedded systems access to human-like visual perception in real-time, which at its core means detecting things, and identifying where they are in physical space. It does this with a combination of traditional machine vision functions (like edge detection and perspective correction), depth sensing, and the ability to plug in pre-trained convolutional neural network (CNN) models for complex tasks like object classification, pose estimation, or hand tracking in real-time.

So how is it used? Practically speaking, the OAK-D Lite is a USB device intended to be plugged into a host (running any OS), and the team has put a lot of work into making it as easy as possible. With the help of a downloadable application, the hardware can be up and running with examples in about half a minute. Integrating the device into other projects or products can be done in Python with the help of the DepthAI SDK, which provides functionality with minimal coding and configuration (and for more advanced users, there is also a full API for low-level access). Since the vision processing is all done on-board, even a Raspberry Pi Zero can be used effectively as a host.

There’s one more thing that improves the ease-of-use situation, and that’s the fact that support for the OAK-D Lite (as well as the previous OAK-D) has been added to a software suite called the Cortic Edge Platform (CEP). CEP is a block-based visual coding system that runs on a Raspberry Pi, and is aimed at anyone who wants to rapidly prototype with AI tools in a primarily visual interface, providing yet another way to glue a project together.

Earlier this year we saw the OAK-D used in a system to visually identify weeds and estimate biomass in agriculture, and it’s exciting to see a new model being released. If you’re interested, the OAK-D Lite is available at a considerable discount during the Kickstarter campaign.

OpenCV Spreads Smart Camera Joy To See Ideas Come To Life

January 14, 2021 by Roger Cheng 4 Comments

Do you have a great application for computer vision, but couldn’t spare the cost of hardware needed to build it? Or perhaps you just need a deadline to pull you away from endless doom scrolling? Either way, the OpenCV team wants you to enter their OpenCV AI Competition 2021 and they’re willing to pitch in hardware to make it happen.

This competition is part of OpenCV’s 20th anniversary celebration, and the field of machine vision has changed a lot in those two decades. OpenCV started within Intel harnessing power of their high end CPUs, but today the excitement is around specialized acceleration hardware for vision processing. Which is why OpenCV put their support and lent their name to the OpenCV AI Kit (OAK) Kickstarter we covered a few months ago. Since then, the hardware was produced and starting to arrive in project backer’s hands. (Barring pandemic-related shipping restrictions…)

This shiny new hardware is the competition’s focus. Phase one solicits team proposals for putting an OAK-D’s power to novel use. University teams may have up to ten members, general teams are limited to four. Each team’s geographic home will put them in one of six global regions. Proposals must be submitted by January 27th, 2021. By February 11th, judges will select the best twenty-five general and ten university team proposals from each region, and every member of the team gets an OAK-D unit to turn their idea into reality by phase two deadline of June 27th. That’s up to 1,200 OAK-D modules available to anyone who can convince the judges they have a great idea and they are capable of bringing it to fruition. Is that you? Of course it is!

Teams will also receive additional resources such as an allotment of cloud compute credits to train their models, and naturally all tutorials and sample code released as part of OAK Kickstarter. No explicit resource for project team organization is mentioned, but of course our own Hackaday.io is available to support you. Best of luck to everyone who enters and we look forward to seeing all the projects this contest will bring to life.

OAK Vision Modules Help You See The Forest And The Trees

August 9, 2020 by Roger Cheng 13 Comments

OpenCV is an open source library of computer vision algorithms, its power and flexibility made many machine vision projects possible. But even with code highly optimized for maximum performance, we always wish for more. Which is why our ears perk up whenever we hear about a hardware accelerated vision module, and the latest buzz is coming out of the OpenCV AI Kit (OAK) Kickstarter campaign.

There are two vision modules launched with this campaign. The OAK-1 with a single color camera for two dimensional vision applications, and the OAK-D which adds stereo cameras for that third dimension. The onboard brain is a Movidius Myriad X processor which, according to team members who have dug through its datasheet, have been massively underutilized in other products. They believe OAK modules will help the chip fulfill its potential for vision applications, delivering high performance while consuming low power in a small form factor. Reading over the spec sheet, we think it’s fair to call these “Ultimate Myriad X Dev Boards” but we must concede “OpenCV AI Kit” sounds better. It does not provide hardware acceleration for the entire OpenCV library (likely an impossible task) but it does cover the highly demanding subset suitable for Myriad X acceleration.

Since the campaign launched a few weeks ago, some additional information have been released to help assure backers that this project has real substance. It turns out OAK is an evolution of a project we’ve covered almost exactly one year ago that became a real product DepthAI, so at least this is not their first rodeo. It is also encouraging that their invitation to the open hardware community has already borne fruit. Check out this thread discussing OAK for robot vision, where a question was met with an honest “we don’t have expertise there” from the OAK team, but then ArduCam pitched in with their camera module experience to help.

We wish them success for their planned December 2020 delivery. They have already far surpassed their funding goals, they’ve shipped hardware before, and we see a good start to a development community. We look forward to the OAK-1 and OAK-D joining the ranks of other hacking friendly vision modules like OpenMV, JeVois, StereoPi, and AIY Vision.

Neural Network Smartens Up A Security System

July 3, 2019 by Lewin Day 11 Comments

It’s all well and good having a security camera recording all the time, but that alone can’t sound the alarm in the event of a crime. Motion sensing is of limited use, often being triggered by unimportant stimuli such as moving shadows or passing traffic. [Tegwyn☠Twmffat] wanted a better security system for the farm, and decided that neural networks would likely do the trick.

The main component of the security system is a Raspberry Pi fitted with a camera and a Movidius Neural Compute Stick. This allows the Raspberry Pi to run real-time object identification on video. The Raspberry Pi is programmed to raise the alarm if it detects humans approaching, but ignores the family dog and other false targets. In the event of a detection, the Raspberry Pi sends a signal over LoRa to a base station, which sounds an alarm. The pitch of the alarm increases the closer the target gets to the camera, thanks to some simple code with bounding boxes.

It’s a nifty way to create an intelligent security system, and all the more impressive for being entirely constructed from off-the-shelf parts and code. Neural networks have become increasingly useful; they can even tell when your cat wants to go outside. Video after the break.

Continue reading →

Raspberry Pi Camera With Smarts — Cloud Or Local?

March 7, 2019 by Al Williams 20 Comments

[Mark West] gave an interesting presentation at last year’s GOTO Copenhagen conference. He shows how he took a simple Raspberry Pi Zero webcam and expanded it with AI. He actually added the intelligent features in two different ways: on in the Amazon cloud and another using the Intel Modvidius NCS USB stick directly connected to the USB. You can see the video below.

Local motion detection uses some open source software. You simply configure it using a text file and it even handles the video streaming. However, at that point, you just have a web camera — not amazing, nor very cost effective. However, you get a lot of false alarms with the motion detection software. A random cat walking past, clouds, trees, or even rain would push [Mark] an email and after 250 alert e-mails a day, [Mark] decided to make something better.

Continue reading “Raspberry Pi Camera With Smarts — Cloud Or Local?” →

Solar Pi Cluster Scours Internet For Nudes

July 6, 2018 by Tom Nardi 50 Comments

There seems to be a universal truth on the Internet: if you open up a service to the world, eventually somebody will come in and try to mess it up. If you have a comment section, trolls will come in and fill it with pedantic complaints (so we’ve heard anyway, naturally we have no experience with such matters). If you have a service where people can upload files, then it’s a guarantee that something unsavory is eventually going to take up residence on your server.

Unfortunately, that’s exactly what [Christian Haschek] found while developing his open source image hosting platform, PictShare. He was alerted to some unsavory pictures on PictShare, and after he dealt with them he realized these could be the proverbial tip of the iceberg. But there were far too many pictures on the system to check manually. He decided to build a system that could search for NSFW images using a trained neural network.

The nude-sniffing cluster is made up of a trio of Raspberry Pi computers, each with its own Movidius neural compute stick to perform the heavy lifting. [Christian] explains how he installed the compute stick SDK and Yahoo’s open source learning module for identifying questionable images, the aptly named open_nsfw. The system can be scaled up by adding more Pis to the system, and since it’s all ARM processors and compute sticks, it’s energy efficient enough the whole system can run off a 10 watt solar panel.

After opening up the system with a public web interface where users can scan their own images, he offered his system’s services to a large image hosting provider to see what it would find. Shockingly, the system was able to find over 3,000 images that contained suspected child pornography. The appropriate authorities were notified, and [Christian] encourages anyone else looking to search their servers for this kind of content to drop him a line. Truly hacking for good.

This isn’t the first time we’ve seen Intel’s Movidius compute stick in the wild., and of course we’ve seen our fair share of Raspberry Pi clusters. From 750 node monsters down to builds which are far more show than go.

Neural Networks… On A Stick!

April 24, 2018 by Al Williams 14 Comments

They probably weren’t inspired by [Jeff Dunham’s] jalapeno on a stick, but Intel have created the Movidius neural compute stick which is in effect a neural network in a USB stick form factor. They don’t rely on the cloud, they require no fan, and you can get one for well under $100. We were interested in [Jeff Johnson’s] use of these sticks with a Pynq-Z1. He also notes that it is a great way to put neural net power on a Raspberry Pi or BeagleBone. He shows us YOLO — an image recognizer — and applies it to an HDMI signal with the processing done on the Movidius. You can see the result in the first video, below.

At first, we thought you might be better off using the Z1’s built-in FPGA to do neural networks. [Jeff] points out that while it is possible, the Z1 has a lower-end device on it, so there isn’t that much FPGA real estate to play with. The stick, then, is a great idea. You can learn more about the device in the second video, below.

Continue reading “Neural Networks… On A Stick!” →