You can find more details about the models for classification, volume estimation, and segmentation here
You can find the client-side here

Nutritional Analysis Server

This repository contains the server side code for an application that estimates the volume of food in images, identifies the type of food, and provides a detailed nutritional analysis. The server is implemented in Node.js, with a series of API endpoints that handle image uploads, process them through machine learning models, and manage the lifecycle of the image files both locally and in Google Cloud Storage.

Overview

The server code is organized around Express.js middleware and a series of endpoints for managing and processing uploaded images. In particular:

• CORS (Cross-Origin Resource Sharing) is enabled to allow for cross-domain access.
• The /uploads route serves static files from an 'uploads' directory.
• Google Cloud Storage is configured for storing and retrieving segmented images.
• Multer middleware is used for handling multipart/form-data, primarily used for file uploads.
• Endpoints are provided for uploading images, retrieving processing results, retrieving and deleting files, and editing data.

Usage

You can start the server locally by running node server.js, assuming you've installed all the necessary dependencies (npm install). The server will start on port 5000, or on the port specified by the PORT environment variable.
It is also hosted on a GCP VM, so you can access it using the client-side without running the server on your local machine.

• I don't keep the models running idly on Google Kubernetes Engine to save on costs.

Endpoints

• POST /api/upload: Accepts an image file for upload and processing.
• GET /api/result/:requestId: Returns the result of processing the image associated with the given request ID.
• GET /api/images: Returns a list of all files in the 'uploads' directory.
• GET /api/deleteAll: Deletes all files in the 'uploads' directory and in the Google Cloud Storage directory.
• DELETE /api/delete/:filename: Deletes the specified file from the 'uploads' directory and from the Google Cloud Storage.
• POST /api/edit: Accepts a new name and updated data, and returns modified nutritional information.

Dependencies

• Node.js and npm
• Express.js for server routing.
• Multer for handling file uploads.
• Google Cloud Storage for storing and retrieving images.
• Axios for making HTTP requests.
• Cors for enabling CORS.
• dotenv for environment variable management.