How to convert torch.tensor to base64 image?

Question

Here is my tensor:

import torch
from torchvision import transforms

content:
tensor([[[[0.8939, 0.8700, 0.8458,  ..., 0.7610, 0.7093, 0.6909],
                                     .
                                     .
                                     .
          [0.4880, 0.5192, 0.5957,  ..., 0.8569, 0.9148, 0.9186]]]])

I want to convert this torch.Tensor to base64. I've tried this but I can not convert PILImage to base64 too.

img = transforms.ToPILImage(content)

How can I do this?

Answer 1

You should first convert your image to bytes to encode afterward your image to base64 image

import torch
from torchvision import transforms
import base64

tensor = torch.randn(3, 512, 512)
transform = transforms.ToPILImage()
pil_image = transform(tensor)

# Convert the PIL image to bytes
image_bytes = pil_image.tobytes()

# Encode the bytes to base64
base64_string = base64.b64encode(image_bytes).decode()
print(base64_string)

Output:

00Hbhc4yZEZnw/VlIyHF6OH2VRAXeRdYiH82tBgSBoocDAB+cycknk3c5G9ZmdbHNu90whgDK49cjELZ6uJCzqYxVclXbJRc0Nb2AgN9XWULFOK+ubWJAyNbAgNZRvmwocwJN4603