PIT: Position-Invariant Transform for Cross-FoV Domain Adaptation (ICCV 2021)

Getting started

---

Clone the repo:

git clone https://github.com/sheepooo/pit-Position-Invariant-Transform/
cd pit-Position-Invariant-Transform

Requirements:

• Python >= 3.6 (numpy, itertools, argparse)
• pytorch >= 0.4.1

To test the PIT function, you can run:

python PIT_tensor.py

The images in "test_images" folder will be PITed and RPITed, and be saved in the same folder.

To PIT all images in a folder, you can run:

python pit_images_in_root_folder.py --fovx 'YourFovx' --root_path 'YourImageFolderName'

either fovx or fovy is enough (both is ok, too). NOTICE: this code would change the images in the root folder directly, so you may need to back up the original images.

To PIT annotations for object detection (XML file in "Pascal VOC" format, as shown in the "test_annotations" folder), you can run:

python pit_annotations.py --fovx 'YourFovx' --root_path 'YourAnnotationFolderName'

either fovx or fovy is enough (both is ok, too). NOTICE: this code would change the annotations in the root folder directly, so you may need to back up the original annotations.