Downsampling fundus images for retinal vessel segmentation reduces thin-vessel sensitivity by up to 15.8 points while Dice remains stable, with optimal processed widths varying by native dataset resolution.
These alterations occur exclusively in the thinnest vessel branches, structures that are often only one to two pixels wide even at native acquisition resolution
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Effect of Input Resolution on Retinal Vessel Segmentation Performance: An Empirical Study Across Five Datasets
Downsampling fundus images for retinal vessel segmentation reduces thin-vessel sensitivity by up to 15.8 points while Dice remains stable, with optimal processed widths varying by native dataset resolution.