Vision AI is a flexible and powerful tool to help developers to bring CV/ML solutions for their Mobile Android projects.
The library reduces the amount of boilerplate code for creating an infrastructure around ML such as - Retrieving video frames and converting it into arrays of color data, Normalizing data, Down/Upscaling of bitmaps, etc - and helps developers be focused on main ML part of the app. With that VisionAI provides simple and understandable API for new ML algorithms and "Processing Steps".
The tool makes it easy to analyze video streams or separate frames and apply ML effects for it. Just in several lines of code!
Example - applying Artistic Style Transferring on a video stream directly on Android Device:
Apache 2.0. See the LICENSE file for details.