Skip to content
←Return to Projects
Case Study

AR Smart Mirror

Real-time interactive AR virtual try-on system designed for retail fashion.

The Problem

In-store clothing fitting is slow, creates long queues in fitting rooms, and increases garment wear. Retailers need an instant, hygienic, and engaging digital mirror that lets shoppers visualize outfits without undressing.

My Role

Built solo with AI-assisted development — computer vision pipeline design, MediaPipe pose detection integration, Canvas AR overlay engine, FastAPI backend services, Docker multi-container topology, and Cloudflare Zero Trust ingress.

Official Website ↗Mobile Controller ↗Architecture Docs ↗Try-On Pipeline ↗

Highlights

  • Sub-30ms real-time body tracking and landmark extraction using MediaPipe Pose
  • Dynamic Canvas AR garment warping and affine transformation for realistic clothing drape
  • FastAPI backend with PostgreSQL catalog and Redis session pub/sub synchronization
  • Mobile controller pairing allowing customers to browse catalogs and trigger try-ons from their smartphones
  • Production-grade Docker Compose architecture with Nginx reverse proxy and Cloudflare Tunnel

Stack

Python 3.11FastAPIMediaPipeHTML5 CanvasPostgreSQL 15Redis 7DockerNginx

Constraints

  • Hard real-time latency target (<33ms per frame / 30 FPS) to avoid motion sickness and visual lag.
  • Zero customer biometric persistence — video feeds must process in memory without saving raw footage to disk.
  • Compatibility with standard consumer hardware (e.g. RTX 3060) without requiring specialized depth cameras.

System Architecture

Capture & Client
Webcam Feed
MediaPipe Pose (Client A)
Mobile Controller (Client B)
Edge & Ingress
Cloudflare Tunnel
Nginx Reverse Proxy
SSL Termination
Application Services
FastAPI Gateway (:8000)
Garment Overlay Service
WebSocket Pub/Sub
Data & Storage
PostgreSQL 15 (Catalog/Users)
Redis 7 (Session Sync)
Local Volume Storage
Telemetry
Prometheus
Grafana Monitoring (:3000)

Key Trade-offs

The decisions worth defending — what I chose, what I turned down, and why.

Pose tracking execution model

Chose

Client-side MediaPipe Pose on Canvas API with GPU acceleration

Rejected

Server-side video streaming and frame-by-frame inference

Server inference introduces network jitter, high bandwidth overhead, and privacy concerns streaming raw camera feeds over the network.

Garment deformation approach

Chose

Real-time 2D affine mesh deformation keyed to skeletal anchor landmarks

Rejected

Full 3D physics-based volumetric cloth simulation

3D cloth simulation cannot sustain 30+ FPS on edge kiosk hardware; 2D mesh triangulation provides fluid response with low computational cost.

Remote store interconnect

Chose

Cloudflare Tunnel over outbound HTTPS

Rejected

Direct port forwarding and dynamic DNS on store routers

Retail store Wi-Fi and ISP firewalls prevent inbound port forwarding; Cloudflare Tunnels punch through NAT securely without public IPs.

What I'd Do Differently

An honest retrospective — the stuff I'd change with more time, more users, or a second pass.

  1. 1Implement WebGPU shaders for cloth lighting and shadow blending to improve realism under dynamic store lighting.
  2. 2Add an offline fallback queue for store analytics so telemetry buffers locally during internet interruptions.
  3. 3Optimize landmark smoothing filters (OneEuro filter) to eliminate micro-jitter when shoppers stand at edge distances.

Computer Vision & Pose Tracking Shell

Benchmark the real-time AR try-on pipeline (status, camera init, pose detect, mesh align -shirt, latency bench).

icelegends-interactive-shell ~ [armirror]
History: ↑ / ↓
cv-engine@ar-mirror:~/pipeline$

Technical Deep-Dive

Architecture, specifications, and implementation details.

16 — Staff Operations Guide

#Daily Morning Startup

  1. Power on the PC — press button, wait 30 seconds
  2. Open Docker Desktop — double-click icon on desktop
  3. Open Terminal — right-click desktop → "Open in Terminal"
  4. Type: docker compose up -d → press Enter
  5. Wait until all 6 containers show green in Docker Desktop
  6. Open browser → go to http://localhost
  7. Test: stand in front — check "POSE: LOCKED" appears

#Adding a New Garment

  1. Go to: http://localhost/admin
  2. Login with admin password (ask manager for this)
  3. Click "Add Garment"
  4. Fill in:
    • Name (e.g. "Black Bomber Jacket")
    • Type: tee / hoodie / jacket / dress / coat
    • Color (e.g. "black")
    • Size range (e.g. "S-XL")
  5. Upload PNG file (transparent background)
  6. Click Save — item appears in mirror within 30 seconds

#Evening Shutdown

  1. Open Terminal
  2. Type: docker compose down → press Enter
  3. Close Docker Desktop (right-click tray icon → Quit)
  4. Power off PC normally via Windows

#Quick Troubleshooting

ProblemSolution
Mirror won't loadCheck Docker is running. Run: docker compose up -d
Camera black screenReload page (F5). Check webcam USB is connected
Body not detectedStand 1m from screen. Check lighting. Reload page
New garment missingWait 30 seconds. Refresh the browser page
Remote access brokenCheck PC is on. Run: cloudflared tunnel run ar-mirror

#Emergency

Call Popa Matei: +40 730 155 075

~ End of Document ~