ValueError: tile cannot extend outside image

[vnyk]

What did you do?

I am trying to crop an image
Sorry, I am not able to provide a source image

A_img = Image.fromarray(A_img[y1:y2, x1:x2, :])

What did you expect to happen?

To get the properly cropped image without any bugs

What actually happened?

File "/home/developer/workspace/dance/few-shot-vid2vid/data/fewshot_pose_dataset.py", line 185, in crop_person_region
    A_img = Image.fromarray(A_img[ys:ye, xs:xe, :])
  File "/home/developer/anaconda3/envs/fewshot-dev/lib/python3.7/site-packages/PIL/Image.py", line 2670, in fromarray
    return frombuffer(mode, size, obj, "raw", rawmode, 0, 1)
  File "/home/developer/anaconda3/envs/fewshot-dev/lib/python3.7/site-packages/PIL/Image.py", line 2613, in frombuffer
    return frombytes(mode, size, data, decoder_name, args)
  File "/home/developer/anaconda3/envs/fewshot-dev/lib/python3.7/site-packages/PIL/Image.py", line 2546, in frombytes
    im.frombytes(data, decoder_name, args)
  File "/home/developer/anaconda3/envs/fewshot-dev/lib/python3.7/site-packages/PIL/Image.py", line 829, in frombytes
    d.setimage(self.im)
ValueError: tile cannot extend outside image

What are your OS, Python and Pillow versions?

• OS: Ubuntu 16.04.6
• Python: 3.7
• Pillow: 6.1.0

[radarhere]

Without an image, this is difficult to debug. I mean, I can construct a script that runs the line you have provided, using one of our test images, and it runs without a problem.

from PIL import Image
import numpy

x1, x2 = 20, 100
y1, y2 = 20, 100
im = Image.open('Tests/images/hopper.jpg')
A_img = numpy.asarray(im)

A_img = Image.fromarray(A_img[y1:y2, x1:x2, :])

[radarhere]

I do find that ValueError: tile cannot extend outside image is raised if I set the dimensions to zero.

Image.fromarray(A_img[50:50, 50:50, :])

Is that the case with your code?

[radarhere]

At the moment, the simplest explanation is that the dimensions that you're trying to crop are wrong. If they're zero, or negative, then I can replicate your error.

If you're able to provide values for x1, x2, y1 and y2 to show how you're cropping the image, that could be helpful. If you're able to let us know the format of your image (JPG, PNG, TIFF, etc.) that would be helpful.

Closing, as without more information, there isn't much to do here.

[ye-yng]

Going to necro this issue with some information. I also encountered this issue when cropping images with bounding boxes detected through AWS Rekognition. The error only happens when negative coordinates are passed to the cropped image.

AWS Rekognition documentation actually has a note documenting the return values for the bounding box: https://docs.aws.amazon.com/rekognition/latest/APIReference/API_BoundingBox.html

[radarhere]

Hi. If you're using the same code in this issue, then I ask you to consider the following.

The following works when x2 is greater than x1 and y2 is greater than y1,

from PIL import Image
import numpy

x1, x2 = 0, 1
y1, y2 = 0, 1
im = Image.open('Tests/images/hopper.jpg')
A_img = numpy.asarray(im)

a = A_img[y1:y2, x1:x2, :]
print(a)
A_img = Image.fromarray(a)

printing

[[[20 21 67]]]

Pillow can load that array without a problem.

If x2 and y2 are the same as x1 and y1, then there is the error.

from PIL import Image
import numpy

x1, x2 = 1, 1
y1, y2 = 1, 1
im = Image.open('Tests/images/hopper.jpg')
A_img = numpy.asarray(im)

a = A_img[y1:y2, x1:x2, :]
print(a)
A_img = Image.fromarray(a)

printing

[]

If x2 and y2 are greater than x1 and y1, then there is the error.

from PIL import Image
import numpy

x1, x2 = 2, 1
y1, y2 = 2, 1
im = Image.open('Tests/images/hopper.jpg')
A_img = numpy.asarray(im)

a = A_img[y1:y2, x1:x2, :]
print(a)
A_img = Image.fromarray(a)

printing

[]

So in those last two examples, Pillow is being passed an empty array. Unless you'd like Pillow to show a (0, 0) image, there's nothing to do, as Pillow isn't being given any image data.

If you're not using the same code as this issue, then please create a new issue with a self-contained example.