比较与tf.image.pad_to_bounding_box的功能差异
tf.image.pad_to_bounding_box
tf.image.pad_to_bounding_box(
image,
offset_height,
offset_width,
target_height,
target_width
)
更多内容详见tf.image.pad_to_bounding_box。
mindspore.dataset.vision.c_transforms.Pad
class mindspore.dataset.vision.c_transforms.Pad(
padding,
fill_value=0,
padding_mode=Border.CONSTANT
)
使用方式
TensorFlow:对图像各边进行填充,入参为上、左边框像素填充数和预期输出图像高、宽,像素填充值为零。
MindSpore:对图像各边进行填充,入参为各边像素填充数、填充值和填充模式。
代码示例
# The following implements Pad with MindSpore.
import numpy as np
import mindspore.dataset as ds
from mindspore.dataset.vision import Border
image = np.random.random((28, 28, 3))
result = ds.vision.c_transforms.Pad((2, 2, 5, 5), 0, Border.CONSTANT)(image)
print(result.shape)
# (35, 35, 3)
# The following implements pad_to_bounding_box with TensorFlow.
import tensorflow as tf
image = tf.random.normal((28, 28, 3))
result = tf.image.pad_to_bounding_box(image, 2, 2, 35, 35)
print(result.shape)
# (35, 35, 3)