The Multiverse Next Door

The author has recently developed an app that uses Machine Learning (ML) to translate bootleg comic books. Despite the growing availability of multilingual comics, there is still a lack of tools for reading them easily. This project aims to introduce new archaeology in comics fandom and provide a simple tool for deciphering translated comics. The app is built using Next.js 10 web framework and relies on Amazon's Amplify CLI and DataStax's Stargate APIs for Apache Cassandra database. The ML functions include handwriting recognition, which is one of the earliest Computer Vision tasks. The CV and NLP models used are often better than 90% accurate, but human correction is still required. The app focuses on creating a solid user experience for editing ML-annotated data while optimizing storage for continuous training and data analysis.