BIOS Hackathon BIOS Hackathon
← Back to gallery

RuntimeTerror

StudyHub

Problem statement: Handwritten + Digital Note Fusion Engine Updated Nov 22, 2025 11:53 IST Upwotes: 1
StudyHub thumbnail

Project overview

Students today juggle a messy mix of information—handwritten lecture notes in notebooks, scattered PPTs, random screenshots, and stacks of PDFs. While each piece is valuable, they rarely connect into a single, usable source of truth. Instead, knowledge remains fragmented across formats and devices. As a result: • Students waste hours searching for “that one slide” or flipping through half-filled notebooks • Revision becomes inefficient because insights are scattered rather than structured • Connections between lecture notes, research papers, and classroom slides are easily missed • Valuable personal notes often remain trapped on paper, inaccessible for digital use Even those who try to stay organized face challenges: • Manual digitization of notes is time-consuming and error-prone • Existing OCR tools capture text but fail to preserve meaning or context • There is no seamless way to merge handwritten content with digital material into a single knowledge base There is a clear need for a solution that: • Uses OCR to capture and digitize handwritten notes with accuracy • Aligns these notes with digital slides, textbooks, and research papers to create a unified view • Generates query able summaries, turning fragmented material into a dynamic, searchable “living textbook” • Supports semantic alignment, allowing cross-referencing between lecture highlights and deeper academic sources • Helps students revise smarter, not harder, by making their entire knowledge ecosystem coherent and accessible Such a platform could transform how students’ study by fusing personal learning with digital resources. Instead of scattered notes and disconnected files, learners would finally have a structured, interactive knowledge hub built from everything they already use.

Inspiration

It was inspired by the real struggle students face managing scattered learning materials—handwritten notes, PPTs, PDFs, screenshots, and textbooks—making revision slow, unorganized, and disconnected. The idea comes from wanting a unified, intelligent system that fuses all these sources into one coherent, searchable study hub.

What it does

It captures handwritten notes, extracts text, aligns them with digital slides/PDFs, and generates unified summaries so students get a single, searchable knowledge hub.

How we built it

Using OCR for handwritten text, text-matching and semantic similarity models to align notes with slides, an LLM for summarization, and a lightweight web interface to present fused content.

Individual contributions

Each team member handled a part: • OCR + preprocessing • Semantic alignment + embeddings • LLM summarization workflow • Frontend UI for unified notebook • Backend integration + APIs

Challenges

Messy handwriting, noisy images, aligning handwritten text to exact slide content, keeping inference fast, and merging multiple formats into one timeline.

Accomplishments

Accurate handwritten OCR, smooth fusion of handwritten + digital notes, fast semantic search, clean UI, and a working end-to-end “living textbook” prototype.

Learnings

How to optimize OCR, handle multimodal text pipelines, improve semantic similarity, manage context windows, and design a user-friendly knowledge interface.

Next steps

Add voice notes, auto-tagging, concept maps, spaced-repetition flashcards, collaboration features, and mobile scanning with real-time syncing.

Back to list