arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2108.04562cs.CV
收藏

开放世界语义分割的深度度量学习

Deep Metric Learning for Open World Semantic Segmentation

The Hong Kong University of Science and Technology(香港科技大学)

Jun Cen, Peng Yun, Junhao Cai, Michael Yu Wang, Ming Liu

摘要:

经典的闭集语义分割网络检测分布外(OOD)对象的能力有限,这对于自动驾驶等安全关键型应用非常重要。增量学习这些带有少量注释的OOD对象是扩大深度学习模型知识库的理想方法。在本文中,我们提出了一个开放世界的语义分割系统,它包括两个模块:(1)一个开放集语义分割模块,检测分布和OOD对象。(2)增量式少样本学习模块,用于逐渐将这些OOD对象合并到其现有知识库中。这个开放世界的语义分割系统表现得像一个人,它能够识别OOD对象,并在相应的监督下逐渐学习它们。我们采用深度度量学习网络(DMLNet)与对比聚类实现开集语义分割。与其他开集语义分割方法相比,我们的DMLNet在三个具有挑战性的开集语义分割数据集上实现了最先进的性能,而无需使用额外的数据或生成模型。在此基础上,进一步提出了两种增量式少样本学习方法,利用面向对象对象的标注对DMLNet进行渐进式改进。

英文摘要:

Classical close-set semantic segmentation networks have limited ability to detect out-of-distribution (OOD) objects, which is important for safety-critical applications such as autonomous driving. Incrementally learning these OOD objects with few annotations is an ideal way to enlarge the knowledge base of the deep learning models. In this paper, we propose an open world semantic segmentation system that includes two modules: (1) an open-set semantic segmentation module to detect both in-distribution and OOD objects. (2) an incremental few-shot learning module to gradually incorporate those OOD objects into its existing knowledge base. This open world semantic segmentation system behaves like a human being, which is able to identify OOD objects and gradually learn them with corresponding supervision. We adopt the Deep Metric Learning Network (DMLNet) with contrastive clustering to implement open-set semantic segmentation. Compared to other open-set semantic segmentation methods, our DMLNet achieves state-of-the-art performance on three challenging open-set semantic segmentation datasets without using additional data or generative models. On this basis, two incremental few-shot learning methods are further proposed to progressively improve the DMLNet with the annotations of OOD objects.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑