HIGHLIGHTS
- who: Xiyao Fu from the Purdue University, United States have published the article: 3D bi-directional transformer U-Net for medical image segmentation, in the Journal: (JOURNAL)
- what: To compensate for the loss of feature resolution brought by transformers, the authors propose 3D transformer UNet (3DTU), which employs a hybrid CNN-transformer architecture to leverage both detailed highresolution spatial information from CNN features and the global context encoded by the new 3D bi-directional transformer module. The authors show that such a design allows the framework to preserve the advantages of self-attention mechanisms and . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.