Custom Image segmentation has different results on different model loads Pixellib

Question

I am doing custom image segmentation using PixelLib.

import pixellib
from pixellib.instance import custom_segmentation

segment_image = custom_segmentation()
segment_image.inferConfig(num_classes= 1, class_names= ["BG", "road"])
segment_image.load_model("/content/mask_rcnn_model.143-0.199986.h5")
segment_image.segmentImage("/content/FullTrainingData/Warlick.PNG", show_bboxes=False, output_image_name="a0000sample_out" + str(i) +".PNG")

The segmentation looks good on my machine, but running it on another machine with the same h5 file the result is far worse

The result differs each time I reload the model. Any reason this could be happening?

bug reports go on https://github.com/ayoolaolafenwa/PixelLib — Christoph Rackwitz, Oct 11 '21 at 19:42
do you think this is a bug? I think it's a problem with my code, not the package if I had to guess — GCK, Oct 11 '21 at 19:53

score 1 · Answer 1 · answered Oct 11 '21 at 20:24

1

adding these imports

import pixellib
from pixellib.tune_bg import alter_bg

causes it to work. No idea why though

answered Oct 11 '21 at 20:24

GCK

63
1
5

Custom Image segmentation has different results on different model loads Pixellib

1 Answers1

Linked