That Label's got Style: Handling Label Style Bias for Uncertain Image Segmentation

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

48 Downloads (Pure)


Segmentation uncertainty models predict a distribution over plausible segmentations for a given input, which they learn from the annotator variation in the training set. However, in practice these annotations can differ systematically in the way they are generated, for example through the use of different labeling tools. This results in datasets that contain both data variability and differing label styles. In this paper, we demonstrate that applying state-of-the-art segmentation uncertainty models on such datasets can lead to model bias caused by the different label styles. We present an updated modelling objective conditioning on labeling style for aleatoric uncertainty estimation, and modify two state-of-the-art-architectures for segmentation uncertainty accordingly. We show with extensive experiments that this method reduces label style bias, while improving segmentation performance, increasing the applicability of segmentation uncertainty models in the wild. We curate two datasets, with annotations in different label styles, which we will make publicly available along with our code upon publication.
Original languageEnglish
Title of host publicationProceedings of Eleventh International Conference on Learning Representations
Number of pages19
Publication date2023
Publication statusPublished - 2023
EventEleventh International Conference on Learning Representations - Kigali Convention Centre, Kigali , Rwanda
Duration: 1 May 20235 May 2023
Conference number: 11


ConferenceEleventh International Conference on Learning Representations
LocationKigali Convention Centre
Internet address


Dive into the research topics of 'That Label's got Style: Handling Label Style Bias for Uncertain Image Segmentation'. Together they form a unique fingerprint.

Cite this