The problem InsightO solves

It recognises text from pics, docs etc. using OCR and then summarises it, saving user's time and efforts. Also recognised text can be exported to a text file. It pre-processes images before OCR'ing image enabling it to work for blurred images, images with not so clear background and also for skewed images. (Images which are not 90 degree up-down, which are taken at some angle). If user wants, user can also input text directly for summarizing it.

Challenges we ran into

1. OCRing image
2. Pre-proessing image to handle various cases
3. Modelling Machine learning models to summarize general text
4. Designing UI using PyQt5