Skip to content

Latest commit

 

History

History
37 lines (25 loc) · 1.22 KB

README.md

File metadata and controls

37 lines (25 loc) · 1.22 KB
Raw

learn.ai Rust Backend

Handles part of the Learn.ai backend. Responsible for calculating engagement scores for the largest face in each image.

Overview

  • Sends binary image via POST request endpoint
  • Is sent through the Azure Face API
  • Parses all face detection data to Rust structs
  • Calculates an engagement score from many different features

Engagement Score calculation

Engagement score Head yaw normalization Head pitch normalization Emotion normalization graph Emotion normalization

Building

Before you are able to run the server, you must create a file called config.rs and add this code to it:

pub static KEY: &'static str = "<Azure Face API key here>";
pub static ENDPOINT: &'static str = "<Azure Face API endpoint here>";

Afterwards, place the config.rs under the src folder.

Now you use Cargo, the Rust package manager, to build the executable. Once you have installed Cargo, clone this repository, add your config.rs, and run:

cargo build --release

the resulting executable will be under target/release/.