Introduction to POSMARS: WebAR & Gamification solution without app installation for events

Hello everyone,

Today I want to introduce POSMARS – a mobile browser WebAR & Gamification solution that I built, serving interactive activities at events and point of sale.

This idea was sparked when I was still working as a designer for a brand. When running activation campaigns or events, brands loved to create games or interactive AR to attract customers. However, in the past, we had to rely entirely on large platforms like TikTok, SparkAR, or some expensive paid WebAR solutions that came with their logos. Meanwhile, getting customers to download a heavy app of several dozen MBs and register an account just to play a 30-second game was something that almost everyone found tedious and overlooked.

When Meta announced the shutdown of SparkAR, I began to delve deeper into WebAR and realized that it was entirely possible to build WebAR from open-source code to create a QR code scanning experience that automatically opens the camera for playing games and receiving rewards for convenience. Later, when I transitioned to working as a freelance 3D CGI artist (specializing in creating graphic assets to support projects), I finally had time to focus on further developing this solution to support my freelance work, increase the value of the 3D CGI assets I created, and test the market.

This is a service that supports the full setup and operation tailored to each campaign's requirements, not a self-service SaaS model.

---

Main technology solutions on POSMARS

The system is developed with a focus on three main interactive feature groups:

* WebAR (Augmented Reality):

* Image/World Tracking (SLAM): Scan product packaging or place interactive 3D models in real space using the phone's camera.

* Face Filter & Hand Gesture: Face filters or hand gesture recognition (like drawing virtual Tarot cards) using recognition technology through the camera.

* Optimize 3D files from Blender: As a 3D CGI artist used to creating heavy files, I built an automatic tool to optimize and compress .glb files directly exported from Blender, reducing size by up to 90% for faster loading on mobile.

* Camera optimization: The entire WebAR camera flow is processed 100% locally on the device to ensure privacy. The system integrates scripts that automatically detect the User-Agent of Zalo/Facebook/TikTok to guide users to switch to Safari/Chrome to open the external browser with just one tap.

* Gamification (Interactive games):

* Includes familiar game templates such as Lucky Draw, Scratch Card, Product Exploration Quiz, and Catcher Game.

* AI Photo Booth:

* Capture selfies directly and use Generative AI to transform portraits according to preset brand styles (Anime, Cyberpunk...). Photos are automatically configured to be permanently deleted from the cloud after 24 hours to protect user information.

---

### Real-life Case Study

The first real project applying this solution was a campaign run for the face mask product lines of the brand Colorkey, implemented simultaneously at physical event booths and online on Facebook.

The designed experience was to draw Tarot cards using hand gestures (Hand Gesture AR): users scan the QR code, hold their hand in front of the phone camera to swipe and select a virtual Tarot card to receive a trial gift.

* Recorded results: Attracted over 1,200 scanning interactions and collected 1,000+ voluntarily registered leads (Name, Phone Number) from customers.

---

### Operation and Management Procedure

Since it is a complete handover solution, the implementation process is simplified for brands and agencies:

* Interactive Demo Approval: Clients receive a trial link to check the interface and effects directly on their phones before going live.

* Real-time asset changes: Image content, 3D models, and prize rates are updated directly through the Admin Dashboard without needing to intervene in the source code.

* Analytics Tracking: The system provides a separate Analytics Dashboard that real-time updates the number of scans, collected