Teaching Robots to See With Javascript
summary
Shows node.js bindings to opencv and walks through the actual implementation details: wrapping the cv matrix in v8, building with package‑config and ffmpeg, and adding a threshold function live. Explains Viola‑Jones / Haar + integral image math and why cascades are fast. Demonstrates a proof‑of‑concept quadrocopter face‑tracking demo (laggy), notes face detection is near real‑time in C++ but the Node bindings and video decoding add significant latency.
AI-written summary · about
chapters (6)
- 0:00Start
- 0:10Teaching Computers to See with Javascript
- 28:01There are a lot of really great algorithms available in OpenCV
- 29:11OpenCV offers new algorithms too
- 30:29node bindings aren't as scary as you think...
- 59:40Explanation 1. Let's check the general structure of the program
related talks
36mMaking a Robot Eye with JavaScript
56mTensorFlow.js: Make a smart webcam in JS with a pre-trained ML model
34mA Powerful Introduction to Machine Learning with Js
21mHands-Free Coding With Gaze Control in JavaScript
36mNo, Really... Robots and JavaScript?!
26mSelf-Portraits in JavaScript
28mBuilding Motion Controlled Websites in JavaScript
15mInferenceJS: Real-time computer vision in your browser