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
Hi,
All the input images identified result in segments being too small or too big for the image. None of the image/segmentation goes past that check.
Output of stats: imgs found: 164 imgs with illegal annotations: 12 imgs with legal annotations: 152 seg too big: 385 seg too small: 495 seg too close to the edges: 2 seg success: 0
everything else is 0...
What could be the issue?
The text was updated successfully, but these errors were encountered:
Did you use the MSCOCO image corpus without rescaling?
It is possible that some images don't have the correct object size, try using more images, perhaps.
I download the images using coco.download(). Is there any rescaling that needs to be done?
Following your advice, I downloaded a lot more images, and now it is fine.
Hi,
All the input images identified result in segments being too small or too big for the image. None of the image/segmentation goes past that check.
Output of stats:
imgs found: 164
imgs with illegal annotations: 12
imgs with legal annotations: 152
seg too big: 385
seg too small: 495
seg too close to the edges: 2
seg success: 0
everything else is 0...
What could be the issue?
The text was updated successfully, but these errors were encountered: