Say It Out Loud: Spoken Interview Scripts

Project Pitch (60 Seconds)

"I built the Gesture PDF Presenter, an interactive touchless presentation system powered by MediaPipe computer vision, React 18, and PDF.js. Using a standard laptop webcam, it tracks 21 hand landmarks in real time to let speakers advance slides, zoom into diagrams, and activate a virtual laser pointer using natural hand gestures in mid-air—eliminating the need for handheld clickers or touching a keyboard."

Feature Scripts

F1: MediaPipe 21-Landmark Tracking Pipeline

"I built the tracking pipeline using Google's MediaPipe Hands in GestureControl.tsx. The webcam captures video at 30 frames per second, passing each frame to MediaPipe's WebAssembly model. It outputs 21 normalized three-dimensional landmarks corresponding to every knuckle and joint in the hand with sub-15 millisecond latency, providing the mathematical coordinates needed for touchless gesture detection."

F2: Touchless Slide Navigation

"To advance slides reliably, in services/gestures.ts I engineered a directional velocity algorithm. A common issue with computer vision presentations is accidental slide changes when a speaker moves their hands while speaking. My engine tracks wrist landmark velocity over five consecutive frames. Only deliberate swipes moving faster than 400 pixels per second trigger a slide transition, followed by an immediate 500-millisecond lock to prevent double-skipping."