torchvision中transform图片通道

最新推荐文章于 2022-09-04 20:33:34 发布

@呵哈哈

最新推荐文章于 2022-09-04 20:33:34 发布

阅读量147

点赞数

分类专栏： pytorch学习文章标签：计算机视觉 opencv python

本文链接：https://blog.csdn.net/qq_43631083/article/details/125453140

版权

pytorch学习专栏收录该内容

3 篇文章 0 订阅

订阅专栏

from torchvision import transforms
from PIL import Image
import cv2

trans = transforms.ToTensor()

img = Image.open('001.jpg')
print(img.size)                # w, h
img_PIL_tensor = trans(img)
print(img_PIL_tensor.size())   # c, h, w

img = cv2.imread('001.jpg')
print(img.shape)               # h, w, c
img_cv2_tensor = trans(img)
print(img_cv2_tensor.size())   # c, h, w

img = np.zeros([100,200,3])    # h, w, c
img_np_tensor = trans(img)
print(img_np_tensor.size())    # c, h, w   torch.Size([3, 100, 200])