Aller au contenu
Parcours du fondateur42 days

Real-time object detection and three-dimensional tracking for a government mixed-reality experience

Delivered in 42 days and demonstrated live at the State Legislative Assembly.

A mixed-reality "Selfie Point" desktop application for the Lucknow Vidhan Sabha: live camera feed, real-time object detection and positional tracking compositing visitors into mixed-reality content, delivered in 42 days.

A public mixed-reality installation for the Uttar Pradesh Legislative Assembly, operated as a desktop application in a live government environment.

Rôle : Computer-vision and application engineering

Delivered in partnership with Live Pixel and Mizzle Codes.

From start to government demonstration
42 daysFrom start to government demonstrationFounder track record
Detection and compositing, on local hardware
Real timeDetection and compositing, on local hardwareNo cloud round-trips
Public installation at the State Legislative Assembly
LivePublic installation at the State Legislative AssemblyLucknow Vidhan Sabha

Le système

Ce qui a réellement été construit

  • Live camera feed captured and processed in real time
  • YOLO-based object detection on every frame
  • Spatial and positional tracking of detected subjects
  • Mixed-reality composition of live subjects with rendered content
  • Desktop application packaging for on-site operation

Pourquoi c’était difficile

Complexité d’ingénierie

  • Real-time detection and compositing on local hardware, no cloud round-trips
  • A fixed 42-day window ending in a live public demonstration
  • Operating reliably in an uncontrolled, high-footfall environment

Expérience produit

Issu du système livré

The mixed-reality desktop application running on a tablet: a subject tracked inside a three-dimensional wireframe bounding box, with selectable subject thumbnails below
Detection and 3D tracking in the delivered app
The mixed-reality application in a window, showing the Uttar Pradesh Legislative Assembly header, an environments panel and three selectable environment tiles over a live camera background
Environment selection over the live camera feed
Mixed reality desktop application demonstrated at the Uttar Pradesh State Legislative Assembly
Demonstration at the State Legislative Assembly

Technologie

Stack vérifiée

Seules les technologies confirmées dans le code source, la documentation du projet ou le déploiement en production sont listées.

  • Python
  • OpenCV
  • YOLO object detection
  • Real-time image processing
  • Three-dimensional tracking

Livraison

Périmètre de responsabilité

  • Detection and tracking pipeline
  • Mixed-reality composition
  • On-site delivery

Similaires

Systèmes proches

Livraison Techon Pixel

Automatic Crime Detector95.1%

A Django web application that analyses webcam and uploaded video frame by frame with trained YOLO models, captures anomalies with timestamps and sends automated email alerts, with remote web access.

Screens of the Automatic Crime Detector web application: the project landing screen, and a detection view showing a live webcam feed beside a detected-anomaly panel
Résultat client publié

American Way Health$430M+

A production voice-AI operation for American Way Health, built around Bland AI. The engineering contribution here is the production layer that keeps the agent calling, compliant and measurable at 400K+ calls per month.

AW

Construire un système d’une complexité comparable

Réservez une consultation gratuite de 30 minutes, ou écrivez-nous — vous parlerez toujours directement à la personne qui construit votre produit.

Réserver un appel gratuit de 30 minDémarrer un projet

Techon Pixel · info@techonpixel.com