You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
End-to-end retail store intelligence system built for Purplle Tech Challenge 2026. Processes raw CCTV footage using YOLOv8 + ByteTrack to detect visitors, tracks movement across zones, and streams structured events into a Spring Boot REST API. Features real-time metrics,conversion funnel analysis, zone heatmaps,and AI-powered anomaly detection
Real-time Store Intelligence System that transforms CCTV footage into actionable retail insights through computer vision, event processing, and analytics dashboards.
Designed and implemented a near-real-time data warehouse for METRO Pakistan using MESHJOIN algorithm in Java for efficient customer transaction analysis.