Vision Enhancement App

A web-based multimodal visual augmentation prototype for low-vision users. It combines real-time contrast enhancement, AR text overlay, OCR with audio output, and AI-assisted scene description in a single zero-installation browser interface.

This repository accompanies the Master's thesis: Khatab, H. (2026). Visual Augmentation and Multimodal Assistive Technology for People with Low Vision: A Literature-Driven Design Study. Tampere University / Tallinn University / Lusofona University.

Submission state

The state of this repository at thesis submission is preserved as the git tag thesis-submission-v1. To view that exact state:

git checkout thesis-submission-v1

Key dependencies

Pinned versions are in package.json and the lockfile. Submission-state versions:

• opencv-js-wasm 5.0.0-alpha - CLAHE via WebAssembly-compiled OpenCV
• @tensorflow/tfjs 4.22.0 + @tensorflow/tfjs-backend-webgl 4.22.0 - TensorFlow.js runtime for Zero-DCE++ inference
• @tensorflow-models/coco-ssd 2.2.3 - pre-trained object detection
• @xenova/transformers 2.17.2 - Transformers.js loading SmolVLM-256M-Instruct
• tesseract.js 7.0.0 - on-device OCR with LSTM engine

Installation and running

pnpm install
pnpm dev
pnpm build
pnpm test

Repository structure

client/                 Frontend source code (React + TypeScript)
server/                 Backend and API routes (Express + tRPC)
shared/                 Shared types and utilities
drizzle/                Database schema and migration config
patches/                Package patches applied by pnpm
public/                 Static assets

License

MIT License - see LICENSE.

Citing this work

If you use this code, please cite:

Khatab, H. (2026). Visual Augmentation and Multimodal Assistive Technology for People with Low Vision: A Literature-Driven Design Study. Master's thesis, Tampere University / Tallinn University / Lusofona University.

A CITATION.cff file is included for citation tooling.