English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Ensuring reliable confidence scores from deep networks is of pivotalimportance in critical decision-making systems, notably in the medical domain.While recent literature on calibrating deep segmentation networks has led tosignificant progress, their uncertainty is usually modeled by leveraging theinformation of individual pixels, which disregards the local structure of theobject of interest. In particular, only the recent Spatially Varying LabelSmoothing (SVLS) approach addresses this issue by softening the pixel labelassignments with a discrete spatial Gaussian kernel. In this work, we firstpresent a constrained optimization perspective of SVLS and demonstrate that itenforces an implicit constraint on soft class proportions of surroundingpixels. Furthermore, our analysis shows that SVLS lacks a mechanism to balancethe contribution of the constraint with the primary objective, potentiallyhindering the optimization process. Based on these observations, we propose aprincipled and simple solution based on equality constraints on the logitvalues, which enables to control explicitly both the enforced constraint andthe weight of the penalty, offering more flexibility. Comprehensive experimentson a variety of well-known segmentation benchmarks demonstrate the superiorperformance of the proposed approach.

Trust your neighbours: Penalty-based constraints for model calibration