A Magical Gathering

on YouTube ↗29m

" summary

A system identifies Magic the Gathering cards using OpenCV for cropping, libpHash perceptual hashes, SQLite sorting and Ruby promises with a thread pool. The corpus covers about 25,000 unique cards; 1.6 GB of images is downloaded and roughly 26,000 hashes are precomputed on four CPUs after releasing the GVL. Capture uses a webcam and lightbox, it learns from manual confirmations, and common failures are slow focus, similar artwork ambiguity, and no integrated pricing yet.

AI-written summary · about

" chapters (11)

1. 0:00Introduction
2. 7:46Hardware
3. 11:55Promises
4. 13:33Limit promises
5. 15:27Perceptual Hatch
6. 17:27Hash Calculations
7. 18:52Recognition Cropping
8. 20:57Preprocessing
9. 25:02Card Matching
10. 25:16Saving Cards
11. 26:20Challenges

" related talks

• 43mOh, Oh, Oh, It's Magic!Aaron Patterson · Ancient City Ruby 2014
• 41mThe Future of Computer Vision: How Two Rubyists Are Changing The WorldAaron Patterson, Jonan Scheffler · Ruby on Ales 2014
• 26mDetecting and classifying object images using rubyFabio Leandro Janiszevski · RubyConf 2024
• 25mMagic is Ruby, Ruby is Magic!Ryan Bigg · RubyConf AU 2024
• 55mFinding DragonsMario Klingemann · Codemotion Roma 2014
• 38mCreative ClojureKarsten Schmidt · clojureD 2016
• 40mComputer Vision Using Ruby and libJITJan Wedekind · RubyConf 2009
• 23mPostcards from GorbyPuffAja Hammerly · RailsConf 2016