Spaces:
Runtime error
Runtime error
| from __future__ import annotations | |
| from typing import Tuple | |
| import numpy as np | |
| def ensure_uint8(image: np.ndarray) -> np.ndarray: | |
| """ | |
| Convert an image into uint8 representation. | |
| """ | |
| image = np.asarray(image) | |
| if image.dtype == np.uint8: | |
| return image | |
| image = image.astype(np.float32) | |
| if image.max() <= 1.0: | |
| image = image * 255.0 | |
| image = np.clip(image, 0, 255) | |
| return image.astype(np.uint8) | |
| def normalize_image(image: np.ndarray) -> np.ndarray: | |
| """ | |
| Convert image to float32 [0, 1]. | |
| """ | |
| image = ensure_uint8(image) | |
| return image.astype(np.float32) / 255.0 | |
| def chw_to_hwc(image: np.ndarray) -> np.ndarray: | |
| """ | |
| Convert CHW tensor/image layout to HWC. | |
| """ | |
| if image.ndim != 3: | |
| raise ValueError( | |
| f"Expected 3D image, received shape {image.shape}" | |
| ) | |
| return np.transpose(image, (1, 2, 0)) | |
| def hwc_to_chw(image: np.ndarray) -> np.ndarray: | |
| """ | |
| Convert HWC image layout to CHW. | |
| """ | |
| if image.ndim != 3: | |
| raise ValueError( | |
| f"Expected 3D image, received shape {image.shape}" | |
| ) | |
| return np.transpose(image, (2, 0, 1)) | |
| def resize_shape( | |
| height: int, | |
| width: int, | |
| target_size: Tuple[int, int], | |
| ) -> Tuple[int, int]: | |
| """ | |
| Return target spatial dimensions. | |
| """ | |
| if height <= 0 or width <= 0: | |
| raise ValueError("Image dimensions must be positive.") | |
| target_height, target_width = target_size | |
| if target_height <= 0 or target_width <= 0: | |
| raise ValueError("Target dimensions must be positive.") | |
| return target_height, target_width |