AI-Object-Detection-Dashboard

A Streamlit web app that visualizes YOLOv8 detection results in real-time.

Problem Statement

Real-time object detection is a critical component in many computer vision applications such as autonomous systems, surveillance, and industrial automation. However, most existing solutions require complex setup, lack visualization tools, and are not easily deployable across different environments.

The AI Web Dashboard for Object Detection provides an end-to-end solution — allowing users to upload images, run YOLOv8-based object detection, and visualize results directly on an interactive web interface. It bridges the gap between deep learning inference and user-friendly visualization, enabling fast experimentation and deployment.

Tech Stack

Layer	Tools / Frameworks
Frontend	Streamlit (interactive dashboard)
Backend API	FastAPI (RESTful inference service)
Model	YOLOv8 (Ultralytics)
Data Handling	OpenCV, Pillow, NumPy
Deployment	Docker, Docker Compose, AWS EC2
Version Control	Git + GitHub Actions (CI/CD)

Features

🖼️ Image Upload Interface: Upload single or multiple images directly from the dashboard.

🤖 YOLOv8 Integration: Run high-accuracy object detection in real-time.

📊 Result Visualization: Display detection boxes, confidence scores, and class labels.

📈 Performance Metrics: View inference time, FPS, and detection summary charts.

🔗 FastAPI REST Endpoint: /predict route for external integrations and automation.

🐳 Dockerized Deployment: Consistent setup and scalable multi-environment deployment.

⚡ CI/CD Pipeline: Auto-build and deploy via GitHub Actions.

System Architecture

             ┌────────────────────────────┐
             │      User Uploads Image     │
             └──────────────┬──────────────┘
                            │
                    ┌───────▼────────┐
                    │   Streamlit UI  │
                    │ (Dashboard App) │
                    └───────┬────────┘
                            │ REST API Call
                    ┌───────▼────────┐
                    │    FastAPI      │
                    │ (Backend API)   │
                    └───────┬────────┘
                            │
            ┌───────────────▼───────────────┐
            │     YOLOv8 Inference Engine    │
            │ (Detection + Postprocessing)   │
            └───────────────┬───────────────┘
                            │
                    ┌───────▼────────┐
                    │  Results JSON   │
                    │ + Annotated Img │
                    └────────────────┘

Demo Screenshots

(To be added later)

Interface	Description
Dashboard UI	Upload and visualize detection results in real-time
Detection Output	Bounding boxes and confidence scores
Metrics View	Detection stats and performance charts

Setup & Installation

Clone the repository

git clone https://github.com//AI-Object-Detection-Dashboard.git cd AI-Object-Detection-Dashboard

Create virtual environment

python -m venv venv source venv/bin/activate # On Mac/Linux venv\Scripts\activate # On Windows

Install dependencies

pip install -r requirements.txt

Run FastAPI backend

uvicorn app.main:app --reload

Run Streamlit dashboard

streamlit run app/frontend.py

🐳 Docker Deployment

Build and launch using Docker Compose

docker-compose up --build

Access the app:

Streamlit Dashboard → http://localhost:8501

FastAPI Docs → http://localhost:8000/docs

Future Enhancements

📹 Add real-time video inference via webcam or RTSP stream.

🧩 Support model switching (YOLOv8, SSD, Faster R-CNN).

🌐 Include batch image uploads and export of annotated results.

🔐 Implement user authentication and role-based access.

📊 Integrate Prometheus metrics for inference monitoring.

🤝 Contributing

Contributions are welcome! Fork the repository and open a pull request with a detailed explanation of your enhancements or fixes.