让不确定像素也参与训练，提升半监督分割精度
Uncertainty-Participation Context Consistency Learning for Semi-supervised Semantic Segmentation
- 引入语义反向传播策略，利用不确定区域的潜在信息
- 在PASCAL VOC和Cityscapes上达到当前最优性能
- 适合需要减少标注依赖的图像分割研究者
半监督语义分割因其能缓解对大量标注数据的依赖而受到广泛关注。然而，现有一致性正则化方法仅使用预测置信度超过固定阈值的高置信像素进行训练，未能充分挖掘网络中不确定像素区域的监督潜力。为此，本文提出不确定性参与上下文一致性学习（UCCL）方法，以探索更丰富的监督信号。具体而言，首先设计语义反向传播更新（SBU）策略，充分利用不确定像素区域的知识，使模型能够从这些区域学习一致的像素级语义信息；此外，提出类别感知知识调控（CKR）模块，促进不同增强视图间类别级语义特征的一致性，强化编码器中类别级语义信息的学习。在两个公开基准数据集上的实验结果表明，所提方法取得了当前最优性能。代码已开源：https://github.com/YUKEKEJAN/UCCL。
原文摘要 · Abstract (English)
Semi-supervised semantic segmentation has attracted considerable attention for its ability to mitigate the reliance on extensive labeled data. However, existing consistency regularization methods only utilize high certain pixels with prediction confidence surpassing a fixed threshold for training, failing to fully leverage the potential supervisory information within the network. Therefore, this paper proposes the Uncertainty-participation Context Consistency Learning (UCCL) method to explore richer supervisory signals. Specifically, we first design the semantic backpropagation update (SBU) strategy to fully exploit the knowledge from uncertain pixel regions, enabling the model to learn consistent pixel-level semantic information from those areas. Furthermore, we propose the class-aware knowledge regulation (CKR) module to facilitate the regulation of class-level semantic features across different augmented views, promoting consistent learning of class-level semantic information within the encoder. Experimental results on two public benchmarks demonstrate that our proposed method achieves state-of-the-art performance. Our code is available at https://github.com/YUKEKEJAN/UCCL.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。