Computer Vision Tagging Utility

This project provides a web-interface for managing and tagging images for T2I/T2V training. The interface allows for easy creation of directories, uploading of images, cropping, letter-boxing, and tagging of images.

While it is in a functional state, it is still a work in progress and may not work 100% as expected. If you encounter any issues, please report them in the Issues tab.

Installation

Currently, the target OS is Linux. It may work on Windows, but it has not been thoroughly tested. To install, follow these steps:

git clone git@github.com:nendotools/cv-tag-ui.git
cd cv-tag-ui
yarn setup

You will be prompted to decide if you want to install with GPU support. Entering n will install dependencies to run with CPU tagging, which can be slow, but should support most use-cases. Entering y will install dependencies to run with GPU support, which requires an NVIDIA GPU and CUDA installed.

After installation, you can start the server with:

yarn serve

Feature Roadmap

• General
  • Better mobile support
  • Handle multiple connected clients (prevent stacking tagging requests)
  • Image grid layout
  • Image sorting
  • Image filtering
  • Recent Project menu
• Images
  • Image upload
  • Image cropping
  • Image letter-boxing
  • Filter images by tag
  • Copy Image
  • Move Image
  • Delete Image
  • Cropping
    • Cropping tool
    • Fixed aspect ratios {1:1, 4:3, 3:4}
    • Letter-boxing {1:1, 4:3, 3:4}
    • remove background
• Tagging
  • Image auto-tagging
  • Individual Image tagging
  • Tag Merging (combining similar tags)
  • Model Tag automation
  • Bulk tagging (add/remove from all images)
  • Recognition Model selection
• Directories
  • Image directory creation
  • Image directory deletion
  • Image directory renaming
  • Image directory moving
  • Kohya project provisioning
    • Create project structure
    • Default config JSON
    • Image directory structure
    • Tag analytics