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

Noisy label problems are inevitably in existence within medical imagesegmentation causing severe performance degradation. Previous segmentationmethods for noisy label problems only utilize a single image while thepotential of leveraging the correlation between images has been overlooked.Especially for video segmentation, adjacent frames contain rich contextualinformation beneficial in cognizing noisy labels. Based on two insights, wepropose a Multi-Scale Temporal Feature Affinity Learning (MS-TFAL) framework toresolve noisy-labeled medical video segmentation issues. First, we argue thesequential prior of videos is an effective reference, i.e., pixel-levelfeatures from adjacent frames are close in distance for the same class and farin distance otherwise. Therefore, Temporal Feature Affinity Learning (TFAL) isdevised to indicate possible noisy labels by evaluating the affinity betweenpixels in two adjacent frames. We also notice that the noise distributionexhibits considerable variations across video, image, and pixel levels. In thisway, we introduce Multi-Scale Supervision (MSS) to supervise the network fromthree different perspectives by re-weighting and refining the samples. Thisdesign enables the network to concentrate on clean samples in a coarse-to-finemanner. Experiments with both synthetic and real-world label noise demonstratethat our method outperforms recent state-of-the-art robust segmentationapproaches. Code is available at https://github.com/BeileiCui/MS-TFAL.

Rectifying Noisy Labels with Sequential Prior: Multi-Scale Temporal Feature Affinity Learning for Robust Video Segmentation