HIGHLIGHTS
- who: Yan Wen et al. from the Laboratory of Vision Engineering (LoVE), School of computer science, University of Lincoln, DQ, UK have published the research: Rethinking the transfer learning for FCN based polyp segmentation in colonoscopy, in the Journal: (JOURNAL)
- what: It is common to adopt random initialisation for those segmentation task-specific networks with customised layers or modules, where the main focus would be the design of new networks to lean more representative features. In the experiments, the binary label is used for training and evaluation of polyp patch classification, while for the segmentation . . .
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.