hwc-ndarray-letterbox

Letterbox an HWC ndarray image to fit the target width and height while updating the homogeneous transformation matrix.

Commonly used in computer vision pipelines (e.g., YOLO object detection), where you need to reverse-map coordinates such as bounding boxes from the preprocessed image back to the original image space.

Usage

import cv2
import numpy as np
from hwc_ndarray_letterbox import hwc_ndarray_letterbox

# Load an image (HWC ndarray)
image = cv2.imread('input.jpg')
current_homogeneous_transformation_matrix = np.eye(3)

# Desired output size
target_width = 640
target_height = 640

# Perform letterbox resize with matrix tracking
(
    letterboxed_image,
    homogeneous_transformation_matrix,
) = hwc_ndarray_letterbox(
    image,
    current_homogeneous_transformation_matrix,
    target_width,
    target_height,
)

print('Homogeneous transformation matrix:', homogeneous_transformation_matrix)

# To map points from letterboxed image back to original:
# inverse_homogeneous_transformation_matrix = np.linalg.inv(homogeneous_transformation_matrix)
# original_point = inverse_homogeneous_transformation_matrix @ np.ndarray([x, y, 1])

Contributing

Contributions are welcome! Please submit pull requests or open issues on the GitHub repository.

License

This project is licensed under the MIT License.