Packages included in requirements.txt. Create a virtual environment and install:

python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

python3 -m venv venv
venv\Scripts\Activate.ps1
pip install -r requirements.txt

1. Change into the directory task1:

cd ./task1

2. Run task1.py:

python ./task1.py

This will output predicted angles, mean squared error, accuracy, and elapsed time to the console.

1. Change into the directory task2:

cd ./task2

2. Run task2.py:

python ./task2.py --test_directory_path ./test/images --train_directory_path ./train/png --sampling_levels 3 4

This will write prediction annotations and images to ./predict/annotations and ./predict/images respectively based on the templates in ./train/png and test images in ./test/images for sampling levels 3 and 4. You may wish to run sampling prediction in parallel in two separate shells for speed.

1. Change into the directory task2:

cd ./task2

2. Run evaluate.py:

python ./evaluate.py --pred_path_root ./predict/annotations/3 --gt_path_root ./test/annotations

This will output evaluation scores for the predictions in ./predict/annotations/3 against the ground truth values in ./test/annotations.

1. Change into the directory task3:

cd ./task3

2. Run task3.py:

python ./task3.py --train_image_directory ./train_images/ --test_image_directory ./TestWithoutRotations/images/ --test_annotation_directory ./TestWithoutRotations/annotations/ --additional_test_image_directory ./Task3AdditionalTestDataset/images/ --additional_test_annotation_directory ./Task3AdditionalTestDataset/annotations/

This will output the TP, FP, FN and recall for each image. You can change the output type with the boolean SHOW and EVAL variables at the beginning of task3.py.