Skip to main content
Artificial Intelligence

Return to Question

Formatting and grammar
Source Link
edit approved Jul 14, 2022 at 13:32
Kroshtan
edit approved Jul 14, 2022 at 13:32
Kroshtan
  • 259
  • 1
  • 10

I'm interestedwondering why many model architectures usinguse binary mask reconstruction via CNN networkfor segmentational CNNs, butand not the regression of mask polygon coordinates? If I'm right manyMany object detectors usinguse regression forto find coordinates of bounding boxes.

I'm interested why many model architectures using binary mask reconstruction via CNN network, but not the regression of mask polygon coordinates? If I'm right many object detectors using regression for find coordinates of bounding boxes.

I'm wondering why many model architectures use binary mask reconstruction for segmentational CNNs, and not regression of mask polygon coordinates? Many object detectors use regression to find coordinates of bounding boxes.

Source Link
Dmitry Sokolov
Dmitry Sokolov
  • 103
  • 6

Why instance segmentation architectures using reconstruction masks but not regression?

I'm interested why many model architectures using binary mask reconstruction via CNN network, but not the regression of mask polygon coordinates? If I'm right many object detectors using regression for find coordinates of bounding boxes.