{Gang Yu Pengwan Yang Tao Hu Chiliang Zhang Yadong Mu Cees G. M. Snoek}

Abstract

Few-shot learning is a nascent research topic, motivated by the fact that traditional deep learning methods require tremen- dous amounts of data. The scarcity of annotated data becomes even more challenging in semantic segmentation since pixel- level annotation in segmentation task is more labor-intensive to acquire. To tackle this issue, we propose an Attention- based Multi-Context Guiding (A-MCG) network, which con- sists of three branches: the support branch, the query branch, the feature fusion branch. A key differentiator of A-MCG is the integration of multi-scale context features between sup- port and query branches, enforcing a better guidance from the support set. In addition, we also adopt a spatial atten- tion along the fusion branch to highlight context information from several scales, enhancing self-supervision in one-shot learning. To address the fusion problem in multi-shot learn- ing, Conv-LSTM is adopted to collaboratively integrate the sequential support features to elevate the final accuracy. Our architecture obtains state-of-the-art on unseen classes in a variant of PASCAL VOC12 dataset and performs favorably against previous work with large gains of 1.1%, 1.4% mea- sured in mIoU in the 1-shot and 5-shot setting.