-
-
Login page
-
Dashboard
-
slot booking
-
Group discussion Room
-
Results
-
Downloadable file
Inspiration
Group discussions are widely used in academics and recruitment, but their evaluation is often subjective, biased, and inconsistent. I wanted to create a system that ensures fair, objective, and scalable assessment using AI instead of relying solely on human judgment.
What it does
The project is an AI-powered Group Discussion Evaluation Platform that analyzes participants in real time. It evaluates gaze, attention, confidence, and participation using computer vision and generates objective scores along with personalized feedback reports.
How we built it
I developed the frontend using React.js and implemented the backend with FastAPI. MediaPipe and OpenCV were used for real-time face and gaze analysis. WebRTC enabled multi-user video interaction, while Firebase handled authentication, database, and hosting. The system processes live video input to generate performance metrics.
Challenges we ran into
I faced challenges in implementing accurate gaze detection and handling multiple users in real time. Ensuring smooth video communication using WebRTC and managing synchronization between frontend and backend was complex. Deployment without paid services was also a major challenge.
Accomplishments that we're proud of
I successfully built a real-time AI-based evaluation system that reduces human bias. The platform can analyze multiple participants simultaneously and generate automated reports with actionable insights. Integrating AI with live video discussion was a significant achievement.
What we learned
I gained hands-on experience in computer vision, real-time systems, and full-stack development. We learned how to integrate AI models with web applications and handle challenges in deployment, networking, and system optimization.
Log in or sign up for Devpost to join the conversation.