ID Paper

Turn both sides of an ID into a clean, print-ready A4 or Letter PDF directly on your device—no account, no cloud upload, no OCR.

Comment

Inspiration

In many stores, offices, and everyday administrative tasks, people still need to copy both sides of an ID card onto a single sheet of paper.

Existing scanner apps are often designed for general documents, require too many steps, include intrusive ads, or upload images to cloud services. I wanted a faster and more private workflow: capture the front and back, correct the edges, and create a print-ready document in seconds.

What it does

ID Paper turns photos of both sides of an ID card into a clean A4 or US Letter document.

The workflow is simple:

  1. Capture or select the front of the card
  2. Capture or select the back
  3. Automatically detect and correct the card boundaries
  4. Manually adjust the four corners if needed
  5. Arrange both sides on a print-ready page
  6. Export the result as a PDF or image

All image processing is designed to happen locally on the user's device. ID Paper does not require an account, cloud upload, or OCR.

How I built it

I used OpenAI Codex to help plan the architecture, implement the interface, debug image-processing workflows, and improve the mobile user experience.

The project is being developed as a mobile-first web application that can be hosted on Cloudflare and later packaged as an Android app.

The application uses browser-based image processing for edge detection, perspective correction, rotation, brightness adjustment, page layout, and PDF generation.

The code is separated into modules for:

  • Image capture and import
  • Automatic edge detection
  • Manual corner adjustment
  • Perspective correction
  • A4 and US Letter layout
  • PDF and image export
  • Privacy and temporary memory cleanup

Challenges I faced

The biggest challenge was making automatic card detection reliable across different backgrounds, lighting conditions, camera angles, shadows, and reflections.

High-resolution images can also consume a large amount of memory on mobile browsers, so the app needs to resize images carefully without reducing print quality.

Another challenge was designing a workflow that remains useful when automatic detection fails. Instead of blocking the user, ID Paper provides a simple manual four-corner adjustment tool.

What I learned

I learned how to break a real-world problem into smaller modules and use Codex as a development partner rather than relying on a single large prompt.

I also learned that privacy should be part of the technical architecture, not just a policy statement. Keeping ID images on the device reduces both user risk and server infrastructure costs.

What's next

Next, I plan to:

  • Improve automatic edge detection accuracy
  • Test on more Android and iOS devices
  • Add offline PWA support
  • Support additional paper sizes and card formats
  • Package the web app for Google Play
  • Add optional, privacy-friendly monetization after the core experience is stable

Built With

Share this project:

Updates