Virtual Try-On API

📌 Overview

This API allows users to virtually try on clothing. The system takes two images as input:

1. Person Image – The image of the person who wants to try on clothes.
2. Clothing Image – Either an image of a standalone piece of clothing or a person wearing the clothing.

🛠 Process

1. Remove Background – The backgrounds of both images are removed.
2. Head Cropping (if applicable) – If the clothing image is of a person, the head is cropped.
3. Virtual Try-On – The processed images are sent to the Try-On API to generate the final output, showing the person wearing the clothing.

---

Example Results

Here are some example results from our Virtual Try-On API:

Input Images

Left: Person Image | Right: Clothing Image

Target Clothing

Target clothing item used in the try-on

Output Result

Final Result: Person wearing the selected clothing item

🚀 Setup Instructions

1️⃣ Create & Activate Virtual Environment

Before installing dependencies, create a virtual environment:

For Windows:

python -m venv venv
venv\Scripts\activate

For macOS/Linux:

python3 -m venv venv
source venv/bin/activate

---

2️⃣ Install Dependencies

Once the virtual environment is activated, install the required dependencies:

pip install -r requirements.txt

---

3️⃣ Start Redis Server

Redis is required for Celery task management. Run:

redis-server

---

4️⃣ Start FastAPI Server

Run the FastAPI application using:

uvicorn main:app --reload

---

5️⃣ Start Celery Worker

Run the Celery worker to process background tasks:

PYTHONPATH=$(pwd) celery -A tasks worker --loglevel=info --pool=threads

---

📡 API Endpoints

1️⃣ Process Images

Endpoint:

POST /process-two-images/

Description:
Uploads two images and starts processing them.

Request Example (Multipart Form-Data):

{
    "base_image": <person_image.jpg>,
    "target_image": <clothing_image.jpg>
}

Response:

{
    "task_id": "12345abcde",
    "message": "Processing started"
}

---

2️⃣ Check Task Status

Endpoint:

GET /task-status/{task_id}

Description:
Checks the processing status of an image processing task.

Response Example (In Progress):

{
    "task_id": "12345abcde",
    "status": "PENDING",
    "result": null
}

Response Example (Completed):

{
    "task_id": "12345abcde",
    "status": "SUCCESS",
    "result": {
        "images": [
            {"url": "https://yourserver.com/segmented_image.jpg"}
        ]
    }
}

🎯 Results Showcase

Key Features Demonstrated

• ✅ Accurate clothing placement
• ✅ Natural body pose preservation
• ✅ High-quality image output
• ✅ Realistic fabric rendering
• ✅ Seamless integration with body shape

---

📞 Contact & Support

For any questions or support, please open an issue in this repository.