Custom Vision Model in minutes running and learning on any Edge with Brain Builder SDK
The Neurala® Brain Builder SDK™ enables organizations to build Custom Vision applications in minutes using Neurala’s Brain Builder SaaS platform, and deploy them on any Compute Edge for inference and further learning.
The Brain Builder SDK is based on Neurala’s award-winning Lifelong DNN technology, the World’s 1st Deep Learning tech able to learn (not only infer…!) on any Compute Edge.
How does Brain Builder SDK work? Very simple! You can build your custom vision model on a browser running Brain Builder by just uploading a few images, and with a few lines of code you can embed the Custom Vision model in any Edge – and continue learning after deployment!
The following diagram shows the typical workflow for development with Brain Builder SDK:
Brain Builder SDK enables Custom Vision model developed with Brain Builder SaaS platform to be embedded in any device with a processor running a Unix/Linux kernel. The SDK enables to both run inference on such devices – e.g., industrial and security cameras, smart phones, drones, robots, etc. – and enables to also perform additional learning directly on the device, without the need to ping the cloud.
We support C++ for Unix, Objective C for iOS, and Java for Android.
The Brain Builder SDK architecture is shown below.
Brain Builder SDK is hardware agnostic and can support edge deployments on lightweight processors such as those found on industrial IoT devices, drones, robots, and other cameras.
Representative examples of hardware and performance:
Try the power of Edge Learning with Neurala’s Brain Builder Companion App!
The app allows users to:
- train Brains in the cloud and test in the field by using their Device
- train new classes directly at the Edge or augment existing ones
- build proof of concept demos in minutes
- help customers connect the dots
- get prototyping done in minutes!
