Saved in:
Bibliographic Details
Main Authors: Zhang, Zhongwen, Boykov, Yuri
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2507.01721
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913922955608064
author Zhang, Zhongwen
Boykov, Yuri
author_facet Zhang, Zhongwen
Boykov, Yuri
contents We consider weakly supervised segmentation where only a fraction of pixels have ground truth labels (scribbles) and focus on a self-labeling approach optimizing relaxations of the standard unsupervised CRF/Potts loss on unlabeled pixels. While WSSS methods can directly optimize such losses via gradient descent, prior work suggests that higher-order optimization can improve network training by introducing hidden pseudo-labels and powerful CRF sub-problem solvers, e.g. graph cut. However, previously used hard pseudo-labels can not represent class uncertainty or errors, which motivates soft self-labeling. We derive a principled auxiliary loss and systematically evaluate standard and new CRF relaxations (convex and non-convex), neighborhood systems, and terms connecting network predictions with soft pseudo-labels. We also propose a general continuous sub-problem solver. Using only standard architectures, soft self-labeling consistently improves scribble-based training and outperforms significantly more complex specialized WSSS systems. It can outperform full pixel-precise supervision. Our general ideas apply to other weakly-supervised problems/systems.
format Preprint
id arxiv_https___arxiv_org_abs_2507_01721
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Soft Self-labeling and Potts Relaxations for Weakly-Supervised Segmentation
Zhang, Zhongwen
Boykov, Yuri
Computer Vision and Pattern Recognition
We consider weakly supervised segmentation where only a fraction of pixels have ground truth labels (scribbles) and focus on a self-labeling approach optimizing relaxations of the standard unsupervised CRF/Potts loss on unlabeled pixels. While WSSS methods can directly optimize such losses via gradient descent, prior work suggests that higher-order optimization can improve network training by introducing hidden pseudo-labels and powerful CRF sub-problem solvers, e.g. graph cut. However, previously used hard pseudo-labels can not represent class uncertainty or errors, which motivates soft self-labeling. We derive a principled auxiliary loss and systematically evaluate standard and new CRF relaxations (convex and non-convex), neighborhood systems, and terms connecting network predictions with soft pseudo-labels. We also propose a general continuous sub-problem solver. Using only standard architectures, soft self-labeling consistently improves scribble-based training and outperforms significantly more complex specialized WSSS systems. It can outperform full pixel-precise supervision. Our general ideas apply to other weakly-supervised problems/systems.
title Soft Self-labeling and Potts Relaxations for Weakly-Supervised Segmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2507.01721