比较与tf.image.resize的功能差异

tf.image.resize

tf.image.resize(
    images,
    size,
    method=ResizeMethodV1.BILINEAR,
    align_corners=False,
    preserve_aspect_ratio=False,
    name=None
)

更多内容详见tf.image.resize

mindspore.dataset.vision.Resize

class mindspore.dataset.vision.Resize(
    size,
    interpolation=Inter.LINEAR
)

更多内容详见mindspore.dataset.vision.Resize

使用方式

TensorFlow:放缩图像至指定大小,支持保留顶角像素点和原始宽高比。

MindSpore:放缩图像至指定大小,当 size 输入为单个整数时,将保持原始宽高比。

代码示例

# The following implements Resize with MindSpore.
import numpy as np
import mindspore.dataset as ds
from mindspore.dataset.vision import Inter

image = np.random.random((28, 28, 3))
result = ds.vision.Resize((14, 14), Inter.BICUBIC)(image)
print(result.shape)
# (14, 14, 3)

# The following implements resize with TensorFlow.
import tensorflow as tf
from tensorflow.image import ResizeMethod

image = tf.random.normal((28, 28, 3))
result = tf.image.resize(image, (14, 14), ResizeMethod.BICUBIC)
print(result.shape)
# (14, 14, 3)