Kohler Mona, Eisenbach Markus, Gross Horst-Michael
IEEE Trans Neural Netw Learn Syst. 2024 Sep;35(9):11958-11978. doi: 10.1109/TNNLS.2023.3265051. Epub 2024 Sep 3.
Humans are able to learn to recognize new objects even from a few examples. In contrast, training deep-learning-based object detectors requires huge amounts of annotated data. To avoid the need to acquire and annotate these huge amounts of data, few-shot object detection (FSOD) aims to learn from few object instances of new categories in the target domain. In this survey, we provide an overview of the state of the art in FSOD. We categorize approaches according to their training scheme and architectural layout. For each type of approach, we describe the general realization as well as concepts to improve the performance on novel categories. Whenever appropriate, we give short takeaways regarding these concepts in order to highlight the best ideas. Eventually, we introduce commonly used datasets and their evaluation protocols and analyze the reported benchmark results. As a result, we emphasize common challenges in evaluation and identify the most promising current trends in this emerging field of FSOD.
人类即使从少数几个例子中也能够学会识别新物体。相比之下，训练基于深度学习的目标检测器需要大量带注释的数据。为了避免获取和注释这些大量数据的需求，少样本目标检测（FSOD）旨在从目标域中少数新类别的目标实例进行学习。在本次综述中，我们概述了FSOD的当前技术水平。我们根据其训练方案和架构布局对方法进行分类。对于每种方法类型，我们描述了一般实现方式以及提高对新类别性能的概念。只要合适，我们就会给出关于这些概念的简短要点，以突出最佳想法。最后，我们介绍常用的数据集及其评估协议，并分析报告的基准结果。因此，我们强调评估中的常见挑战，并确定这个新兴的FSOD领域中最有前景的当前趋势。
