Skip to content

1.0: Initial Release

Pre-release
Pre-release

Choose a tag to compare

View all tags
@AdityaHegde712 AdityaHegde712 released this 15 Nov 12:29
· 1 commit to main since this release
1.0
400ee51

Release 1.0 - Segmentation-to-Bounding Box Converter

Overview

This is the first official release of the Segmentation-to-Bounding Box Converter. The tool is designed to seamlessly convert binary segmentation masks into YOLO-compatible bounding boxes with enhanced efficiency. This release introduces key features like microbox removal, nested box elimination, and the novel MinIoU metric for merging overlapping boxes.

New Features

  • Conversion of binary segmentation masks to YOLO bounding boxes.
  • Removal of microboxes smaller than 0.1% of the image area.
  • Elimination of nested bounding boxes.
  • Merging of overlapping boxes using a custom MinIoU metric.

Installation

  • Tested on Python 3.10.15.
  • Install dependencies using pip install -r requirements.txt.

Usage

Detailed usage instructions are available in the README.md, with examples for running the script and viewing help.

Known Issues

  • Can only work with binary-class, single channel masks at the current release.

Planned Improvements

  • Adding support for multi-class RGB masks.
  • Adding multiprocessing support.

Thank you for using the Segmentation-to-Bounding Box Converter!

Assets 2
Loading