Regularizing the predicted GND shape parameter β with total variation during image translation yields cleaner aleatoric uncertainty maps and small reconstruction gains on medical I2I tasks.
Medical image translation using an edge-guided genera- tive adversarial network with global-to-local feature fusion
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Uncertainty-Aware Regularization for Image-to-Image Translation
Regularizing the predicted GND shape parameter β with total variation during image translation yields cleaner aleatoric uncertainty maps and small reconstruction gains on medical I2I tasks.