label2rgb raises a wrong message when inputs are 3D

[patquem]

Hi,
the 2D-test presented in #4657 doesn't work when extended in 3D.
Is this a bug ?
Thanks,
Patrick

version : scikit-image-0.18.2

import numpy as np
from skimage.color import label2rgb

img = np.random.randint(0, 255, (10, 10, 10), dtype=np.uint8)
labels = np.zeros((10, 10, 10), dtype=np.uint8)
labels[:, 1:3, 1:3] = 1
labels[:, 6:9, 6:9] = 2
color = label2rgb(labels, image=img, bg_color=(0,0,0), bg_label=0)

raise ValueError("image and label must be the same shape") (in skimage\color\colorlabel.py line 162)

[rfezzani]

According to its documentation, label2rgb only supports 2D inputs: label and image must be respectively of shape (M, N) and (M, N, 3). But this can be effectively checked to raise a proper error message.
I modified your issue title to reflect this. Thank you for the report.

[patquem]

thanks for the reply @rfezzani .
I agree with you (doc + issue title).
As skimage.measure.label works whatever the array size, it could be good to have label2rgb working in 3D (or whatever the dimension, if this has sense), couldn't it ?
(+ type:enhancement ??)