https://github.com/waikato-ufdl/wai-annotations-yolo

wai.annotations module for managing YOLO datasets.
https://github.com/waikato-ufdl/wai-annotations-yolo

image-annotation processing python3

Last synced: about 1 month ago
JSON representation

wai.annotations module for managing YOLO datasets.

• Host: GitHub
• URL: https://github.com/waikato-ufdl/wai-annotations-yolo
• Owner: waikato-ufdl
• License: apache-2.0
• Created: 2022-02-23T03:58:20.000Z (almost 3 years ago)
• Default Branch: main
• Last Pushed: 2024-02-02T03:12:34.000Z (11 months ago)
• Last Synced: 2024-11-08T21:17:56.211Z (about 1 month ago)
• Topics: image-annotation, processing, python3
• Language: Python
• Homepage: https://ufdl.cms.waikato.ac.nz/wai-annotations-manual/
• Size: 24.4 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • Changelog: CHANGES.rst
  • License: LICENSE

Awesome Lists containing this project

README

        
# wai-annotations-yolo

wai.annotations module for managing YOLO datasets. 

The manual is available here:

https://ufdl.cms.waikato.ac.nz/wai-annotations-manual/

## Plugins

### FROM-YOLO-OD

Reads image object-detection annotations in the YOLO format

#### Domain(s):

- **Image Object-Detection Domain**

#### Options:

```

usage: from-yolo-od [-I FILENAME] [-i FILENAME] [-N FILENAME] [-n FILENAME] [-o FILENAME]

                    [--seed SEED] [--image-path-rel PATH] [-p] [-l PATH]

optional arguments:

  -I FILENAME, --inputs-file FILENAME

                        Files containing lists of input files (can use glob syntax) (default: [])

  -i FILENAME, --input FILENAME

                        Input files (can use glob syntax) (default: [])

  -N FILENAME, --negatives-file FILENAME

                        Files containing lists of negative files (can use glob syntax) (default: [])

  -n FILENAME, --negative FILENAME

                        Files that have no annotations (can use glob syntax) (default: [])

  -o FILENAME, --output-file FILENAME

                        optional file to write read filenames into (default: None)

  --seed SEED           the seed to use for randomisation (default: None)

  --image-path-rel PATH

                        Relative path to image files from annotations (default: None)

  -p, --use-polygon-format

                        Reads the annotations in polygon format rather than using auto-detection of

                        bbox or polygon format. (default: False)

  -l PATH, --labels PATH

                        Path to the labels file (default: None)

```

### TO-YOLO-OD

Writes image object-detection annotations in the YOLO format

#### Domain(s):

- **Image Object-Detection Domain**

#### Options:

```

usage: to-yolo-od [-c PATH] [-l PATH] [-p] [--annotations-only] [--no-interleave] -o PATH

                  [--split-names SPLIT NAME [SPLIT NAME ...]] [--split-ratios RATIO [RATIO ...]]

optional arguments:

  -c PATH, --labels-csv PATH

                        Path to the labels CSV file to write (default: None)

  -l PATH, --labels PATH

                        Path to the labels file to write (default: None)

  -p, --use-polygon-format

                        Outputs the annotations in polygon format rather than bbox one. (default:

                        False)

  --annotations-only    skip the writing of data files, outputting only the annotation files

                        (default: False)

  --no-interleave       disables item interleaving (splitting will occur in runs) (default: False)

  -o PATH, --output PATH

                        output directory to write images and annotations to (default: None)

  --split-names SPLIT NAME [SPLIT NAME ...]

                        the names to use for the splits (default: [])

  --split-ratios RATIO [RATIO ...]

                        the ratios to use for the splits (default: [])

```