The Oak-1 is an all-in-one machine vision solution. It’s a 4-trillion-operations-per-second AI powerhouse that performs your AI models on-board, so that your host is free to do whatever you need it to do.
Its integrated 12 MegaPixel camera module communicates over an on-board 2.1 Gbps MIPI interface directly to the Myriad X, which ingests this data and performs neural inference on it, returning the results over USB.
Such a data path offloads the host processor from all of this work. In the common use case of object detection from a 12MP image, this means your host is now dealing with a 24 Kbps stream of what the objects are and where they are in the image, instead of a 2.1 Gbps stream of video. So an 87,500 reduction in data your host has to deal with.
And such a reduction means that even on relatively-slow hosts, one can use dozens of Luxonis OAK-1 without burdening the host CPU.
As an example, below is an example of running MobileNet-SSD:
OAK-1 + Raspberry Pi: 50+FPS, 0% RPi CPU Utilization
NCS2 + Raspberry Pi: 8FPS, 225% CPU Utilization
If a board needs code or communicates somehow, you're going to need to know how to program or interface with it. The programming skill is all about communication and code.
Skill Level: Experienced - You will require a firm understanding of programming, the programming toolchain, and may have to make decisions on programming software or language. You may need to decipher a proprietary or specialized communication protocol. A logic analyzer might be necessary.
See all skill levels
If it requires power, you need to know how much, what all the pins do, and how to hook it up. You may need to reference datasheets, schematics, and know the ins and outs of electronics.
Skill Level: Competent - You will be required to reference a datasheet or schematic to know how to use a component. Your knowledge of a datasheet will only require basic features like power requirements, pinouts, or communications type. Also, you may need a power supply that?s greater than 12V or more than 1A worth of current.
See all skill levels
Based on 2 ratings:
1 of 1 found this helpful:
This is a valuable camera with many sources of use. Although this is the single lens camera and not the three lens camera, it works well.
Some of the ideas from DepthAI will not work on this camera since it only has one lens. For instance, depth perception is out of the game on this module.
P.S. The camera gets a bit hot but one could solve this with a fan from a microcontroller or Single Board Computer. Oh and 64-bit only. This camera will not work on older ARM processors that are 32-bit, e.g. Sitara Processors armv7l am335x or am5729 for instance.
I know a lot of technology for CNN, this is the easiest to handle and most compact that I know, great product.