unable to detect objects

[patricxu]

I use the following code to do segmentation but all it output is a "bottle"

import pixellib
from pixellib.torchbackend.instance import instanceSegmentation

ins = instanceSegmentation()
ins.load_model("pointrend_resnet50.pkl")
target_classes = ins.select_target_classes(person = True)
ins.segmentImage("image.jpg", show_bboxes=True, segment_target_classes = target_classes, output_image_name="output_image.jpg")

I got the model from PointRend

The outcome I got is

even no objects got segmented on some input images.

The following message is the output from my terminal

[ayoolaolafenwa]

@patricxu Are you sure that this is the output image it produced? I tested the same image and I got the desired result. What pytorch version are you using?

[patricxu]

thanks！ I upgraded torch to 1.9 then fix the issue