You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
An AI-powered proctoring tool for secure online exams. Uses YOLOv5 for object detection, MTCNN for face detection, YAMNet for audio detection, and PyQt5 for the UI. Enforces fullscreen mode, disables shortcuts, and provides real-time monitoring with violation tracking. Scalable and resource-efficient.
A multimodal deepfake detection system that analyzes images, videos, and audio to classify content as "Real" or "Fake." This project integrates advanced machine learning techniques and is designed for educational and research purposes.
Snap your fingers and vanish from your own video in Google Meet - a 100% local Chrome extension that dissolves your camera feed into particles and leaves an empty room behind. No server, no data sent. Developed by JOJIN JOHN.
Wristband that detects a doorbell, a smoke alarm or an appliance beep and alerts deaf and hard-of-hearing users with a different vibration for each. It uses explainable FFT detection in C on an nRF52840.
Do AI-music detectors work on generators they never saw? A leave-one-generator-out study on FakeMusicCaps: 97.5% balanced accuracy on known generators, but only about half of the clips caught from an unseen one.