Intelligent Card Detection for Automatic Documentation & e-Commerce

The goal of this project is to develop a Card Detection for Automatic Documentation solution that will be able to detect all features of a collectible card from a scanned image. Given a large number of design variations in cards, simple OCR cannot provide all available information. We aim to deliver an AI-driven solution deploying learning-based approaches and computer vision algorithms to extract all possible variations of elements found on a card. New card entries will be added to the database for online trading. Web-crawler tools will be developed to search reference information from the Internet to put a price tag on the card. This full-stack framework includes a front-end to back-end routing and task execution pipeline. We anticipate benefits to not only the collectible card online trading businesses, but also Canada technologically and economically.

Faculty Supervisor:

Irene Cheng

Student:

Ankush Goyal;Subho Ghose;Jiaxin Xu;Yu Sun;Jia Wan

Partner:

JACOBB

Discipline:

Computer science

Sector:

Professional, scientific and technical services

University:

University of Alberta