Foilcase Scanner Uses Vision and OCR to Tell Apart Trading Card Printings
The developer's iPhone app Foilcase scans and prices cards from five trading card games, identifying printings by matching artwork with Vision feature prints and reading collector numbers with text recognition. Because alternate arts and parallel rarities often share a collector number and illustration, even the correct number yields only 89.5% top-1 accuracy, so the app asks the user to pick the print rather than guessing. Testing on a Mac harness with 9,903 indexed cards and 400 degraded queries showed that cropping choice and frame sharpness strongly affect accuracy, with a distance ceiling of 0.80 used to reject non-card images. The write-up also describes storage using an append-only ledger and a single daily price file shared by all users.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)
Log in to join the discussion and vote.
Log in