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,thx for your pro. I‘m doing post processing after segmentation,but I do not have good ideas. You mention boundary snap in the paper,but I do not know how to use it to test the effect. Could you help me?
The text was updated successfully, but these errors were encountered:
THX~ Is it that you train a CNN model to detect all counters in the image,and then you pick the counter which most fit the mask you get in foreground branch ?
The counter seems not that accurate in Qualitative results on DAVIS 2016 in your paper. Is it because of video detection or counter-detection model? I'm doing segmentation on pictures. If I use your counter-detection model and snap method, will I receive a better result?
Do you have any good ideas how to refine the mask when we get segmentation mask by network? Or whether there are good ways to improve the final result?
Thank you for your reply. Best wishes~
Hi,thx for your pro. I‘m doing post processing after segmentation,but I do not have good ideas. You mention boundary snap in the paper,but I do not know how to use it to test the effect. Could you help me?
The text was updated successfully, but these errors were encountered: