Recent Advances in Open Set Recognition: A Survey
Recent Advances in Open Set Recognition: A Survey
复制标题
开放集识别的最新进展：一项调查
DOI：
10.1109/tpami.2020.2981604
复制
发表时间：
2021-10-01
影响因子：
23.6
通讯作者：
Chen, Songcan
中科院分区：
文献类型：
--
作者：
Geng, Chuanxing;Huang, Sheng-Jun;Chen, Songcan
在实际的识别/分类任务中，由于受各种客观因素的限制，在训练识别器或分类器时，通常很难收集训练样本来穷尽所有类别。一个更现实的场景是开集识别（OSR），其中在训练时存在对世界的不完整知识，并且在测试期间可以将未知的类提交给算法，这要求分类器不仅要准确地分类可见的类，还要有效地处理不可见的类。本文提供了一个全面的调查，现有的开集识别技术，涵盖各个方面，从相关的定义，表示的模型，数据集，评价标准，算法比较。此外，我们简要分析了OSR和它的相关任务，包括零杆，一杆（少数）识别/学习技术，分类与拒绝选项，等等之间的关系。此外，我们还回顾了开放世界识别，它可以被视为OSR的自然延伸。重要的是，我们强调了现有方法的局限性，并指出了一些有前途的后续研究方向在这一领域。
In real-world recognition/classification tasks, limited by various objective factors, it is usually difficult to collect training samples to exhaust all classes when training a recognizer or classifier. A more realistic scenario is open set recognition (OSR), where incomplete knowledge of the world exists at training time, and unknown classes can be submitted to an algorithm during testing, requiring the classifiers to not only accurately classify the seen classes, but also effectively deal with unseen ones. This paper provides a comprehensive survey of existing open set recognition techniques covering various aspects ranging from related definitions, representations of models, datasets, evaluation criteria, and algorithm comparisons. Furthermore, we briefly analyze the relationships between OSR and its related tasks including zero-shot, one-shot (few-shot) recognition/learning techniques, classification with reject option, and so forth. Additionally, we also review the open world recognition which can be seen as a natural extension of OSR. Importantly, we highlight the limitations of existing approaches and point out some promising subsequent research directions in this field.