The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
CrowdVision AI – Real-Time Crowd Monitoring & Intelligence System
An AI-powered computer vision platform that detects and counts people in real time, monitors crowd density and occupancy, tracks entry and exit movements, and provides actionable crowd intelligence through a live monitoring dashboard.
Overview
CrowdVision AI is an intelligent crowd monitoring and safety analytics platform designed to help organisations monitor public spaces, understand crowd behaviour, and identify potentially unsafe crowd conditions.
The system uses computer vision and machine learning inference to analyse camera feeds and video streams, detect people, estimate occupancy, classify crowd density, and monitor movement across designated zones.
The Problem:
Monitoring crowded environments manually can be difficult, time-consuming, and inefficient. Security personnel may struggle to maintain accurate occupancy counts, monitor multiple zones, and identify unusual crowd conditions quickly.
CrowdVision AI was developed to explore how computer vision and real-time analytics could improve crowd visibility and monitoring.
My Solution:
I developed a computer vision system integrating a pretrained YOLO object-detection model with OpenCV for video processing.
The platform includes:
Real-time people detection and counting.
Crowd occupancy estimation.
Crowd-density classification.
Entry and exit line counting.
Zone-based monitoring.
Incident and anomaly monitoring workflows.
Camera health and tamper monitoring.
Evidence management and reporting.
Telegram alert notifications.
Remote edge camera processing.
Live monitoring dashboard.
Technical Implementation:
Machine Learning & Computer Vision
Python
YOLO / Ultralytics
OpenCV
NumPy
Pandas
Backend:
Python
FastAPI
REST API
Frontend:
React
Vite
JavaScript
Other Technologies
Git
GitHub
Telegram Bot API
JSON/CSV data storage
My Role
Founder & Full-Stack Machine Learning Developer
Nice question. It uses yolo tracking with persistent IDs to follow people across frames rather than counting every detection independently. Also has a short tolerance for missed detections to help maintain occupancy counts. But when people heavily overlap, tracking IDs can...
For entry/exit line counting, I'd test someone pausing on the boundary and stepping back without entering. Do you require a track to move fully between two zones before recording an entry, so small position changes don't become extra visits?
Yes. The counter uses the person's foot position (bottom of the bounding box) and a 14-pixel buffer around the counting line. It only records an entry or exit when the same tracked person moves completely beyond the buffer on the opposite side. If someone pauses on the boundary...
Graete
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
Steer an endless night-flight over a procedural city. NOCTURNE is a night-flight airline for insomniacs. An infinite procedural city passes beneath the wing. Steer with your mouse. Built by Samin Sami.
I built this project to test Steer an endless night-flight over a procedural city. NOCTURNE is a night-flight airline for insomniacs. An infinite procedural city passes beneath the wing. Steer with your mouse. Built by Samin Sami. The core engineering challenge was sustaining uncompromised 60 FPS rendering using Three.js (GLB 3D) while handling continuous user interaction without framework overhead.
WHAT I WAS TESTING:
I focused on two core engineering benchmarks:
Precision implementation of infinite world under continuous interactive user input
Sustained 60 FPS render performance and zero-dependency browser execution
THE TECHNICAL SIDE:
Engineered directly in the browser with Three.js (GLB 3D), Canvas2D, Infinite world. The application runs without third-party frameworks or build-step dependencies, using hardware-accelerated rendering and custom event loops to achieve zero-latency response.
Everything executes in the browser using Three.js (GLB 3D), Canvas2D, Infinite world. No third-party frameworks, no heavy runtime bundles, and zero external tracking.
WHAT I LEARNED:
Balancing the mathematical rigors of infinite world with tactile browser interaction required tuning event throttling and memory allocation to eliminate all garbage collection stutters.
Finding the sweet spot between mathematical precision and frame budget took careful profiling, but the result runs at a steady 60 FPS even on modest hardware.
TRY IT YOURSELF:
The live build is deployed on Netlify. Open it in your browser, test the controls, and observe the render response times.
If you are building high-performance 3D, physics, or creative web systems and want to collaborate, reach out here on Contra.
Anyone wants to see the source code can check it out in my JSFiddle: https://jsfiddle.net/Saminsami/n97a8xw1/1
jsfiddle.net
Nocturne - Samin Sami
Nocturne - Samin Sami - a HTML example you can edit and run in your browser on JSFiddle.
A few pieces of the website we recently created for Kodilion. Com check the complete website at https://kodilion.hr/
Eyewear Ecommerce Website Redesign — AI vs. WaxyWeb
A side-by-side design comparison exploring two approaches to an online eyewear store.
The WaxyWeb redesign focuses on the relationship between product presentation, collection browsing, editorial imagery, and the store’s visual identity.
Design considerations:
Eyewear product presentation
Ecommerce layout and hierarchy
Collection and product discovery
Fashion-focused visual direction
Consistent online shopping experience
For fashion, eyewear, and lifestyle brands, the goal is to create a store that presents products clearly while giving the brand a recognisable visual presence.
Website design and development: www.waxyweb.com
WaxyWeb