Creativity and innovation
We ideated an interaction flow where the users can complete quizzes by selecting the correct answer using their hands, feet, or other assigned body parts via a web camera. Using our system, teachers can quickly create quizzes in three steps: (1) Use the question generator, (2) Select a fitness routine; our system will place the correct answers in the positions that will move all assigned body parts, (3) Share it with the students. We also make the system inclusive by allowing teachers to select the body parts that will be active through the quiz.
Impact on the identified challenge and UN Sustainable Development Goals
Quiz and presentation platforms like Kahoot and Menti have become very popular among the education and corporate community. They are engaging as it keeps the brain cognitively active. But they also keep students and corporate workers seated and limit physical activity. Our system and interaction technique can have a significant impact on Goal 3-Good Health and Well-being, and 4-Quality Education and Goal.
Implementation feasibility
We utilized p5.js and ml5.js to detect body points via camera. We used plain Javascript and HTML for the question generation interface. We also promoted the Question Generation NLP project by Questgen AI to accelerate the adaptation of this interface by quickly creating new quizzes. To show the implementation feasibility, we open-sourced the building blocks of this app in an interactive format.
Question Generation: There are already example question generation examples like @ramsrigouthamg’s QuestGen
Pose Detection: We created a simple example that positions multiple-choice questions and runs PoseNet to select the correct answer: See in P5 Editor
Question Front-End: Teachers can also create their own questions. See in CodePen
Built With
- ml5.js
- p5.js
- python
- tensorflow.js

Log in or sign up for Devpost to join the conversation.