HIGHLIGHTS
- who: Ivan Grubiu0161iu0107 et al. from the Faculty of Electrical Engineering and Computing, University of, Unska, have published the paper: Revisiting Consistency for Semi-Supervised Semantic Segmentation, in the Journal: Sensors 2023, 23, 940. of December/31,/2022
- what: The experiments show clear advantages of_(1) one-way consistency perturbing only the student branch and_(3) strong photometric and geometric perturbations. Different from all previous approaches, the authors explore an exhaustive set of consistency formulations. This approach has been adapted for dense prediction by expressing the discriminator as a segmentation network that produces dense C . . .
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.