Lauzhack 2018 project, made by Ulysse Ramage, Jean Chambras, Julien Malka and Gaspard Peduzzi

Table of Contents


Dependencies :gear:

  • In order to be able to run the project you need python v3.5 and node v10 with npm and pip3
  • You need to create a config.js file at the following path client/src/api/config.js with this content :
export default {
  // Your endpoint for Microsoft Azure
  faceUri: "",
  // Your subscriptions key for this service
  faceKey: "xxxxxxxxxxxxxxxxxxxxxxx",
  // The endpoint of your python server used for handraising recognition
  handUri: ""

Launch the project :rocket:

Start the server project with:

cd server
pip3 install -r requirements.txt

Start the application project in an other window with:

cd client
npm install
npm run start

Motivations :mortar_board:

The goal was to create a very intuitive dashboard in the time available where a speaker (teacher, speaker, etc.) could understand and be informed of the audience's interactions. They are categorized into two categories:

  • The hands raised, allowing to directly conduct polls with this interaction, or to note the participation of people in relation to the timeline of the session
  • People's expressions, analyzing the following feelings in all people at every moment of the experience: anger, contempt, disgust, fear, happiness, neutral, sadness, surprise :relieved: :neutral_face: :worried:


This project uses state of the art methods in the fields of computer vision and machine learning to be able to identify emotions of the students and raising hand gestures.

For the latter, we've been using OpenPose to compute a pose estimation of the students in the room and hence know if they're raising their hand.

Share this project: