Progressive web app decoding hidden information in HQR images
# install dependencies
yarn
# serve with hot reload at localhost:8080
yarn dev
# build for production with minification
yarn build
# build for production and view the bundle analyzer report
yarn build --report
For detailed explanation on how things work, checkout the guide and docs for vue-loader.
- The app does not truely work offline due to a problem with OpenCV
- The marker identification is extremely simple leading to false-positives
- The decoding is time consuming (try using TensorFlow.js)
- OpenCV is consuming a lot of memory and iOS devices can crash after some time
- When an image is uploaded using the 'chose file' input, camera-orientation is not handled