Skip to content

Latest commit

 

History

History
67 lines (48 loc) · 2.14 KB

README.md

File metadata and controls

67 lines (48 loc) · 2.14 KB
Raw

FrameNest

FrameNest is a tool that uses computer vision to archive, and create searchable image databases with descriptive metadata. The type of database it builds streamlines the training process of AI models in sectors such as advanced image recognition.
( I basically built it because I myself needed it )

Screenshot

 

Prerequisites

Make sure you have Node.js and Mongo Database installed and running on your Linux.

Additionally, you could also install MongoDB Compass to view the database, it's basically a GUI for MongoDB. ( although not required if you're comfortable with the command line )

Make sure you have the following installed:

Last but not least, you need an Astica account to use the computer vision API.

 

Installation

  1. Go to the root directory of the project and create a .env file with the following content and replace the values with your own:
ASTICA_KEY=< replace this with your astica computer vision key >
DB_CONNECTION_STRING=< replace this with your mongodb connection string >
  1. Install the dependencies:
npm install && npm fund
  1. Lastly, make db_upload.sh and start.sh executable:
chmod +x db_upload.sh start.sh

 

Usage of the webapp

  1. Run the following command in the root directory of the project:
./start.sh
  1. Open your browser and go to
http://localhost:3000/

⚠️⚠️⚠️ Upload a batch of images by putting them inside a zip folder. Please note that the zip folder should only contain images not nested inside any other folders.

 

OS specific instructions

This installation guide is for Linux users.

 

TODO

  • Add code documentation
  • Add Windows installation guide
  • Add MacOS installation guide