Hands-On One-shot Learning with Python

Object detection in few-shot domains

RepMet: Representative-based metric learning for classification and few-shot object detection (https://arxiv.org/abs/1806.04728) is a few-shot learning object detection method. In this paper, the authors proposed a variant of a feature pyramid network for region proposals of objects, and on top of it, they added a metric-based classifier that classifies proposed regions on the basis of distance from learned class representatives. They also made a contribution to the research community by setting up a benchmark on the ImageNet dataset for the few-shot object detection task.

Similarly, One-shot object detection with co-attention and co-excitation (https://arxiv.org/abs/1911.12529) also works on filtering ...