Grove Vision AI V2 - First Impressions

Описание: Here I explore the Seeed Studio Vision AI V2 module. It is used for image processing and in particular object recognition within an image. It can recognise people, faces animals, vehicles in road traffic etc

I have had V1 of this module for about 18 months and it was easy to deploy and interfaced with processors such as the Xiao ESP-C3. Its only problem was a slow frame rate and poor optics with its embedded camera. Version 2 is quite different - better camera and optics, faster refresh rate and better energy efficiency.

In my first impressions of this module I am going to use existing image recognition software models already embedded within the on-line Seeed Studio Sensecraft tool to upload images to the Vision AI V2. I will demo a face recognition model and in the next video I will go thru all of the steps to create this model using an example from the Roboflow library. The face recognition model worked well with an accuracy of greater than 80% confidence.

In this video I also explore the hardware that is compatible with the Vision AI V2 module including he Xiao range of processors and various Raspberry Pi cameras.

So far I am quite impressed with this hardware and will use it on many future projects such as developing a virtual robot path to navigate on mars and recognise coloured musical bells to play music with a robot arm.

References:
Grove Vision AI V2 at Seeed Studio - Shop ( https://www.seeedstudio.com/Grove-Vis... )
Grove Vision AI V2 at Seeed Studio - wiki (https://wiki.seeedstudio.com/grove_vi... )
Sensecraft AI ( https://seeed-studio.github.io/SenseC... )
Roboflow - universe models ( https://universe.roboflow.com/otc-z2i... )