Polyp segmentation with consistency training and continuous update of pseudo-label

HIGHLIGHTS

• who: Hyun-Cheol Park from the DepartmentGachon University have published the article: Polyp segmentation with consistency training and continuous update of pseudo-label, in the Journal: Scientific Reports Scientific Reports
• what: The authors propose encoder-decoder method that is well suited for the polyp which have varying shape, size and scales.. The authors propose an enhanced consistency regularization method to utilize unlabeled data and encourage the model to perform consistent predictions for the same input under different perturbations.. The authors propose continuous update of pseudo-label generated by the average of teacher-student model to . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or E-mail:

  If you don't have an account, you can create one here.