Is-This-Food is an ML-powered web application that analyzes an uploaded image to determine whether it contains any food items. Using an efficient TensorFlow Lite model, this tool provides a quick and intuitive way to check if a given photo consists of food.

View the Live Demo

• Overview
• Features
• How It Works
• Installation
• Usage
• Contributing
• License
• Acknowledgements

Is-This-Food is designed to provide an easy-to-use interface for detecting food in images. Whether you're curious about your latest photo or want to build a food-related application, this project demonstrates how to leverage machine learning in a user-friendly web app.

• ML-Based Food Detection:
  Uses a pre-trained TensorFlow Lite model (EfficientNet) to determine if an image contains food.

• Simple, Clean Interface:
  A straightforward UI built with HTML, CSS, and JavaScript that makes image upload and result display intuitive.

• Responsive Design:
  Works seamlessly on desktop and mobile browsers.

• Live Demo:
  Try out the app instantly via our live demo hosted on Vercel.

1. Image Upload:
   Users can select an image from their device to analyze.

2. Preprocessing:
   The uploaded image is preprocessed (resized and normalized) to meet the input requirements of the model.

3. Inference:
   The TensorFlow Lite model, based on EfficientNet, processes the image to determine the presence of food.

4. Result Display:
   The app displays the prediction result along with confidence scores, letting users know if food was detected.

For more technical details, refer to the accompanying Jupyter Notebook (data_exploration _01.ipynb) which outlines our model training and exploration process.

To run this project locally, follow these steps:

1. Clone the Repository:

   git clone https://github.com/rohan-patnaik/Is-This-Food.git
   cd Is-This-Food

2. Install Dependencies: If you plan to run the app locally using a simple web server, you may use:

   npm install

   or simply open index.html in your browser for a static preview.

3. Set Up a Local Server (Optional): You can use Live Server in VSCode or run:

   npx http-server

   to serve the project files on a local port.

1. Open the Application:

   • Open index.html directly in your browser or run it via your local server.

2. Upload an Image:

   • Use the provided interface to upload a photo from your device.

3. View Results:

   • The application will process the image and display whether food items are detected, along with any relevant confidence metrics.

4. Experiment:

   • Try different images and see how the model performs on various food and non-food images.

This project is licensed under the MIT License. See the LICENSE file for details.

• TensorFlow Lite & EfficientNet:
  Thanks to the contributors of TensorFlow Lite and EfficientNet for providing powerful tools to run ML models on edge devices.

• Vercel:
  For hosting the live demo and making deployment simple.