But none of these methods can cope with the situation where the novel target data is few. Perez-Rua et al. (2020) introduced the incremental few-shot object detection (iFSD) paradigm, where new classes are made available gradually through different sessions. Perez proposed the ONCE model to ...
本文是一篇关于少量样本增量学习(Few-shot Class-Incremental Learning, FSCIL)的综述,提出了一种新的分类方法,将FSCIL分为五个子类别,并提供了广泛的文献回顾和性能评估,讨论了FSCIL的定义、挑战、相关学习问题以及在计算机视觉领域的应用。 1 介绍 年份:2024 作者:田松松,中国科学院半导体研究所;李璐思,老道明大...
With the rapid development of remote sensing technology and the growing demand for applications, the classical deep learning-based object detection model is bottlenecked in processing incremental data, especially in the increasing classes of detected objects. It requires models to sequentially learn new ...