search

DebeshJha / 2020-CBMS-DoubleU-Net

Official implementation of DoubleU-Net for Semantic Image Segmentation in TensorFlow & Pytorch (Nominated for Best Paper Award (IEEE CBMS))
https://arxiv.org/pdf/2006.04868.pdf
239 stars 69 forks source link

Comparison to ResUNet++ #3

Closed Mano3 closed 4 years ago

Mano3 commented 4 years ago

Hi, you have mentioned in your paper ResUNet++, which is also your work. Also, I've seen that you are working on ResUNet++ + CRF + TTA. Can you make a comparison between the 3 on the presented datasets? Is any one of these models superior to the others? Thanks.

DebeshJha commented 4 years ago

Hi,

Thanks for your query. The combination of ResUNet++, CRF, and TTA is data-dependent. However, I would prefer "ResUNet++, +CRF+ TTA" as they perform well on the cross-dataset or generalizability test.