You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
issue #1691 mentioned training panoptic segmentation models on a custom dataset.
Since we can train the built-in COCO or Cityscapes dataset,
I think we can also train our own custom dataset if its format is COCO or Cityscapes.
Is there any annotation(labeling) tool for panoptic segmentation?
or, if there is no tool,
is there any way to make a custom annotation from the existing Instance & Semantic segmentation dataset?
I think the Panoptic FPN uses the "fused" dataset, registering it with register_coco_panoptic_separated.
In that case, I think we can make a custom panoptic dataset from the existing Instance & Semantic segmentation dataset.
But what if we wanna use Panoptic Deeplab, using register_coco_panoptic?
(register a “standard” version of COCO panoptic segmentation dataset)
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
issue #1691 mentioned training panoptic segmentation models on a custom dataset.
Since we can train the built-in COCO or Cityscapes dataset,
I think we can also train our own custom dataset if its format is COCO or Cityscapes.
Is there any annotation(labeling) tool for panoptic segmentation?
or, if there is no tool,
is there any way to make a custom annotation from the existing Instance & Semantic segmentation dataset?
I think the Panoptic FPN uses the "fused" dataset, registering it with
register_coco_panoptic_separated
.In that case, I think we can make a custom panoptic dataset from the existing Instance & Semantic segmentation dataset.
But what if we wanna use Panoptic Deeplab, using
register_coco_panoptic
?(register a “standard” version of COCO panoptic segmentation dataset)
Beta Was this translation helpful? Give feedback.
All reactions