> ## Knowledge Base Index
> Fetch the complete knowledge base index at: https://helping.alive-pic.com/sitemap.xml
> Use this file to discover available pages before exploring further.
> Pure-Markdown content can be obtained by appending a '.md' suffix to the content URLs listed in the sitemap (without the trailing slash).

# How AR Photo Recognition Works

The ALIVE Guest App uses augmented reality to recognize printed photos and play animated videos on top of them. Here's what's actually happening behind the scenes — and why it feels like magic.

---

## The Short Version

Guest points phone at printed photo → app recognizes the image → animated video appears on top of it in real time.

No QR code needed. No manual input. The app figures out which photo it's looking at automatically.

---

## How the Recognition Works

The ALIVE Guest App uses a continuous matching system:

1. **The app captures frames** from the phone's camera at regular intervals
2. **Each frame is sent to the ALIVE backend**, which compares it against all the photos taken at recent events
3. **When a match is found**, the backend returns the corresponding animated ALIVE video
4. **The app loads the AR content** and begins tracking the physical photo in real time
5. **The animated video plays** directly on top of the tracked image — aligned, scaled, and anchored to the print

This all happens in seconds. The guest just sees the photo come alive.

---

## What Makes It Work Well

* **No setup required** — The matching happens automatically. No booth operator configuration needed.
* **Works with any ALIVE photo** — Every photo captured with the ALIVE Effect is automatically available for AR playback.
* **Dynamic library** — The app doesn't store reference images locally. It downloads them on-the-fly when a match is found, so it's always up to date.
* **Smooth tracking** — Once the app locks onto a photo, the video follows the print smoothly as the guest moves their phone around it.

---

## The Technology

The app uses **AR Foundation** (Apple ARKit on iOS, Google ARCore on Android) for image tracking. When the backend identifies a match:

* The reference image is added to the device's AR image library at runtime
* The AR system tracks that image in 3D space
* A video overlay is positioned and scaled to match the physical print exactly
* The video streams from the cloud — no large downloads needed

---

## Tips for Best Results

* **Good lighting helps** — AR tracking works best in well-lit environments. Dim or uneven lighting can slow down recognition.
* **Hold the photo steady** — A flat, uncrumpled print gives the best tracking. Bent or folded photos may not track as reliably.
* **Full photo visible** — Make sure the entire print is visible in the camera frame. Partially covered photos may not be recognized.
* **Internet required** — The matching and video streaming both need an active internet connection.

| The AR recognition only works with photos taken using the ALIVE Effect. Regular photos without an animated video won't trigger AR playback.

---

## What's Next?

* [Guest Experience — Using the AR App](/en/article/guest-experience-using-the-ar-app-1cv6coi) — What guests see and do
* [Recording and Downloading AR Content](/en/article/recording-and-downloading-ar-content-ohx47u) — Capturing the AR moment