Co-Seg: An Image Segmentation Framework Against Label Corruption
Co-Seg: An Image Segmentation Framework Against Label Corruption
复制标题
DOI：
10.1109/isbi48211.2021.9433790
复制
发表时间：
2021-01
期刊：
影响因子：
--
通讯作者：
Ziyi Huang;Haofeng Zhang;A. Laine;E. Angelini;C. Hendon;Yu Gan
中科院分区：
文献类型：
--
作者：
Ziyi Huang;Haofeng Zhang;A. Laine;E. Angelini;C. Hendon;Yu Gan
受监督的深度学习性能与高质量标签的可用性密切相关。如果直接在噪声数据集上训练，神经网络会逐渐过度拟合损坏的标签，导致测试时性能严重下降。在本文中，我们提出了一种新的深度学习框架，即Co-Seg，用于在包含低质量噪声标签的数据集上协作训练分割网络。我们的方法首先同时训练两个网络来筛选所有样本，并获得具有可靠标签的子集。然后，一个有效的但易于实现的标签校正策略，以丰富的可靠子集。最后，使用更新的数据集，我们重新训练分割网络以确定其参数。在两个有噪声的标签场景中的实验表明，我们提出的模型可以实现与在无噪声标签上训练的完全监督学习所获得的结果相当的结果。此外，我们的框架可以很容易地在任何分割算法中实现，以增加其对噪声标签的鲁棒性。
Supervised deep learning performance is heavily tied to the availability of high-quality labels for training. Neural networks can gradually overfit corrupted labels if directly trained on noisy datasets, leading to severe performance degradation at test time. In this paper, we propose a novel deep learning framework, namely Co-Seg, to collaboratively train segmentation networks on datasets which include low-quality noisy labels. Our approach first trains two networks simultaneously to sift through all samples and obtain a subset with reliable labels. Then, an efficient yet easily-implemented label correction strategy is applied to enrich the reliable subset. Finally, using the updated dataset, we retrain the segmentation network to finalize its parameters. Experiments in two noisy labels scenarios demonstrate that our proposed model can achieve results comparable to those obtained from fully supervised learning trained on the noise-free labels. In addition, our framework can be easily implemented in any segmentation algorithm to increase its robustness to noisy labels.