Skip to content

Latest commit

 

History

History
43 lines (34 loc) · 1.41 KB

README.md

File metadata and controls

43 lines (34 loc) · 1.41 KB
Raw

Rotating Yolo-Type Annotations

Rotating segmentation annotations and images of type Yolo. If you want to rotate images and segmentation annotations, you can use this repo. It also works with bounding boxes.

  • Just works with CCW 90 right now.

before_after.png

How to Use

Edit rotate_segmentation_dataset.py for the dataset path, and run the code. It will create a new dataset folder named with {your_dataset_path}_rotated and its subfolders (train, test, val)/(images/labels).

  • For Bounding Box Datasets: rotate_bbox_dataset.py
  • Notebooks can be examined to see how the codes work.
python rotate_segmentation_dataset.py

python rotate_bbox_dataset.py
  • Also, this generates annotated images for each train/test/val folder with the test_images name.
├───train
│   ├───images
│   │       vertical_1.jpg
│   │       vertical_2.jpg
│   │
│   ├───labels
│   │       vertical_1.txt
│   │       vertical_2.txt
│   │
│   └───test_images
│           vertical_1.jpg
│           vertical_2.jpg

TODO

  • Fix for object detection dataset rotation.
  • Give different colors for every class.
  • Put text on annotation on drawing function.
  • Drawing Segmentation annotations from folder main code.
  • Drawing Bounding Box annotations from folder main code.
  • Changing Rotation Degree.