KRISH09bha/object-detection-kodrish

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

🎯 Object Detection & Counting using YOLOv11 + FastAPI

An object detection and counting system using YOLOv11 (12x.pt) powered by FastAPI, with a clean frontend built using modern web tools.

🖼 Upload an image → ⚡ Get real-time detections → 🔢 See object count → 📤 Download annotated image.

🌐 Live Links

• —🔗 Frontend Web App: https://object-hy5d.vercel.app/
• —🚀 API on Hugging Face: https://krish09bha-object-detection-kodrish.hf.space

📁 Project Structure

.
├── .gitattributes           # Git LFS support for model
├── 12x.pt                   # YOLOv11 model weights (~119MB)
├── app.py                   # FastAPI backend logic
├── Dockerfile               # For containerization
├── requirements.txt         # Python dependencies
└── README.md                # Project documentation

🧠 Model Details

• —Model: YOLOv11
• —Trained weights: 12x.pt
• —Purpose: Object detection and counting
• —Framework: Ultralytics YOLO

⚙️ FastAPI Endpoints

POST /upload-image/

Upload an image to get predictions and object count.

Request (multipart/form-data):

• —file: JPEG or PNG image file

Response (JSON):

{
  "image": "data:image/jpeg;base64,...",
  "object_count": {
    "person": 2,
    "car": 1
  }
}

• —image: Annotated image (base64-encoded)
• —object_count: Class-wise count of detected objects

GET /

Basic root endpoint.

{
  "message": "🎯 Object Detection API for Images and Videos using 12x.pt"
}

💻 Local Setup

Step 1: Clone the Repo

git lfs install
git clone https://huggingface.co/spaces/KRISH09bha/object-detection-kodrish
cd object-detection-kodrish

Step 2: Install Dependencies

pip install -r requirements.txt

Step 3: Start FastAPI App

python app.py

API will be available at: http://localhost:8000

🐳 Run with Docker

docker build -t yolo-fastapi .
docker run -p 7860:7860 yolo-fastapi

🧪 Technologies Used

• —Backend: FastAPI
• —Model: YOLOv11 (Ultralytics)
• —Frontend: Hosted at https://object-hy5d.vercel.app/
• —Deployment: Hugging Face Spaces + Vercel + Git LFS

🛡 Error Handling

• —Warns if 12x.pt is missing
• —Returns 500 if image processing fails

👨‍💻 Developer

Made with ❤️ by Krish Bhagat 🚀 Founder at KodRish Innovation & Solution LLP

📧 Email: bhagatkrish65@gmail.com 🔗 LinkedIn: Krish Bhagat

📄 License

This project is licensed under the MIT License.