Lune

CVPR2026Top-tier venue

From Softmax to Dirichlet: Evidential Learning for Semi-supervised Semantic Segmentation

Huayu Mai, Rui Sun, Yujia Chen, Wangkai Li, Bingzhou Wang, Aibing Li, Zhangyu He, Yuan Wang

Abstract

The critical challenge of semi-supervised semantic segmentation lies in how to fully exploit a large volume of unlabeled data to improve the model's generalization performance for robust segmentation. However, existing softmax scores-based filtering methods tend to be affected by the overconfidence issue in neural networks, leading to the inclusion of incorrect pseudo-labels that negatively impact the training process. In this paper, we propose a novel evidential learning framework to explicitly model the prediction uncertainty for reliable pseudo-label selection. By modeling the distribution of class probabilities using Dirichlet distributions, we obtain principled and improved uncertainty estimates from a distributional perspective. Furthermore, we propose HESS (Hyper-ESS), decoupling the modeling of exclusive and collective evidence for comprehensive evidence perception, to yield more accurate uncertainty estimates. Extensive experiments on three challenging benchmarks demonstrate that integrating HESS into existing semi-supervised semantic segmentation frameworks consistently improves performance, benefiting from more reliable pseudo-label selection. Our work sheds light on the potential of evidential learning in semi-supervised semantic segmentation and opens up new avenues for future research. Code and models will be made available to facilitate future research.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Pixel Contrastive-Consistent Semi-Supervised Semantic Segmentation cites FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence
  • Semi-Supervised Semantic Segmentation via Adaptive Equalization Learning cites FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on33

  1. FixMatch: Simplifying Semi-Supervised Learning with Consistency and ConfidenceKihyuk Sohn, David Berthelot, Nicholas Carlini, Zizhao Zhang et al.NeurIPS 2020 · 5,129 citations
  2. RandAugment: Practical Automated Data Augmentation with a Reduced Search SpaceEkin Dogus Cubuk, Barret Zoph, Jonathon Shlens, Quoc LeNeurIPS 2020 · 4,453 citations
  3. Deep Evidential RegressionAlexander Amini, Wilko Schwarting, Ava Soleimany, Daniela RusNeurIPS 2020 · 777 citations
  4. Semi-Supervised Semantic Segmentation via Adaptive Equalization LearningHanzhe Hu, Fangyun Wei, Han Hu, Qiwei Ye et al.NeurIPS 2021 · 221 citations
  5. Pixel Contrastive-Consistent Semi-Supervised Semantic SegmentationYuanyi Zhong, Bodi Yuan, Hong Wu, Zhiqiang Yuan et al.ICCV 2021 · 210 citations

Related papers

  1. Asymmetric Beta Loss for Evidence-Based Safe Semi-Supervised Multi-Label LearningHao-Zhe Liu, Ming-Kun Xie, Chen-Chen Zong, Sheng-Jun HuangKDD 2024 · 1 citation
  2. Evidential Prototype Learning for Semi-supervised Medical Image SegmentationYuanpeng He, Lijian Li, Tianxiang Zhan, Chi-Man Pun et al.KDD 2025 · 1 citation
  3. Multi-view Evidential Learning-based Medical Image SegmentationChao Huang, Yushu Shi, Waikeung Wong, Chengliang Liu et al.AAAI 2025 · 5 citations
  4. DAW: Exploring the Better Weighting Function for Semi-supervised Semantic SegmentationRui Sun, Huayu Mai, Tianzhu Zhang, Feng WuNeurIPS 2023 · 40 citations
  5. Enhanced Soft Label for Semi-Supervised Semantic SegmentationJie Ma, Chuan Wang, Yang Liu, Liang Lin et al.ICCV 2023 · 55 citations
Dusk over the sea between two cliffs drawn in fine vertical lines