!!! Check the image in devpost for the entire pipeline implementation results. In the video the screen recording was not captured for the last part.

Luddy Hackathon 2026!

A two-stage pipeline that ingests a noisy scanned document image, extracts its text using a convolutional neural network (CNN), and compresses the output using adaptive Huffman encoding, delivered as two communicating microservices.

Built With

Share this project:

Updates