Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

scribble supervision

Scribble supervision is a weakly supervised learning paradigm in computer vision where segmentation models are trained using sparse, hand-drawn line strokes over target objects and background regions instead of exhaustive, pixel-level label masks. This approach significantly lowers the time and expense required for manual data annotation by capturing only a minimal subset of representative pixels across an image. Models trained under scribble supervision typically rely on label propagation, edge-aware loss functions, structural priors, or consistency regularization to infer the remaining unlabeled regions and predict complete, dense segmentation masks for full images.

1 item

CycleMix: A Holistic Strategy for Medical Image Segmentation from Scribble Supervision

CycleMix: A Holistic Strategy for Medical Image Segmentation from Scribble Supervision

Ke Zhang, Xiahai Zhuang

OrganizationsFudan University

Why you should read this

Proposes a weakly-supervised medical image segmentation framework that combines mixup-based scribble augmentation with global and local cycle-consistency regularization to achieve segmentation accuracy comparable to fully-supervised methods.

Curating a large set of fully annotated training data can be costly, especially for the tasks of medical image segmentation. Scribble, a weaker form of annotation, is more obtainable in practice, but training segmentation models from limited supervision of scribbles is still challenging. To address the difficulties, we propose a new framework for scribble learning-based medical image segmentation, which is composed of mix augmentation and cycle consistency and thus is referred to as CycleMix. For augmentation of supervision, CycleMix adopts the mixup strategy with a dedicated design of random occlusion, to perform increments and decrements of scribbles. For regularization of supervision, CycleMix intensifies the training objective with consistency losses to penalize inconsistent segmentation, which results in significant improvement of segmentation performance. Results on two open datasets, i.e., ACDC and MSCMRseg, showed that the proposed method achieved exhilarating performance, demonstrating comparable or even better accuracy than the fully-supervised methods. The code and expert-made scribble annotations for MSCMRseg are publicly available at https://github.com/BWGZK/CycleMix.

Added

2026-09-26