How kollinikhilkumar/ai-powered-smart-mall-management-system works
Project type and purpose
• This is a full-stack AI-driven Smart Mall Management System.
• Backend: Python-based ML pipeline + FastAPI microservice.
• Frontend: React/Vite dashboard for real-time analytics.
• Purpose: ingest camera feeds, detect & recognize faces, compute foot-traffic analytics, and display insights to mall administrators.
File structure
• Top-level:
• Backend/
– Fast-Apis/: FastAPI endpoints (main.py)
– Main/: ML pipeline package (src/components, src/pipeline), entrypoint app.py, artifacts, logs
• Dashboard/: React code (src/pages, src/components, charts), Vite build (dist/)
• setup.py, requirements.txt: Python packaging & dependencies
• README.md: methodology & usage instructions
• Patterns:
• “components” & “pipeline” modules implement a modular ML workflow.
• Fast-API folder serves REST API.
• Dashboard folder is a self-contained frontend application.
• Deploy/config files: package.json, vite.config.js, setup.py, requirements.txt
README for insights
• Four modules:
1. Parallel camera input (multiprocessing)
2. YOLO-v8 face detection
3. Face encoding & comparison
4. Customer analytics
• Usage: run FastAPI, run app.py (pipeline), run React dev/build.
Detail is essential
• Python module as a sub-component if needed (data_ingestion.py, model_trainer.py, etc.).
• How FastAPI imports or invokes parts of the ML package, if applicable.