Object Segmentation gives no output for timelapse data

[zoccoler]

Hi everyone,

I noticed that the Object Segmentation (Tools -> Segmentation Labeling -> Object Segmentation (APOC)) does not generate a new labels layer with the output. I believe this may be a bug 🐛. It also does not throw an error.

Minimal Working Example:

import napari
import numpy as np
image = np.random.randint(low=0, high=255, size=(10, 30, 30, 30))
label_image = np.zeros_like(image)
label_image[5, 10:20, 10:20, 10:15] = 1
label_image[5, 10:20, 10:20, 15:20] = 2
viewer = napari.Viewer()
viewer.add_image(image)
viewer.add_labels(label_image)

Then open Object Segmentation, select the image, select ground truth and click on Run.

I get the following printed output, but no new layer:

train ObjectSegmenter
num images 1
object annotation value 2
features gaussian_blur=1 difference_of_gaussian=1 laplace_box_of_gaussian_blur=1 sobel_of_gaussian_blur=1
depth 2
num trees 100
file C:/Users/mazo260d/Documents/project/ObjectSegmenter.cl
annotation shape (30, 30, 30)
Training done. Applying model...
Applying / prediction done.

Versions

python 3.9
napari 0.4.17
napari-accelerated-pixel-and-object-classification 0.10.2