Adaptive Prototype Learning and Allocation for Few-Shot Segmentation

Prototype learning is extensively used for few-shot segmentation. Typically, a single prototype is obtained from the support feature by averaging the global object information. However, using one prototype to represent all the information may lead to ambiguities. In this paper, we propose two novel modules, named superpixel-guided clustering (SGC) and guided prototype allocation (GPA), for multiple prototype extraction and allocation. Specifically, SGC is a parameter-free and training-free approach, which extracts more representative prototypes by aggregating similar feature vectors, while GPA is able to select matched prototypes to provide more accurate guidance. By integrating the SGC and GPA together, we propose the Adaptive Superpixel-guided Network (ASGNet), which is a lightweight model and adapts to object scale and shape variation. In addition, our network can easily generalize to k-shot segmentation with substantial improvement and no additional computational cost. In particular, our evaluations on COCO demonstrate that ASGNet surpasses the state-of-the-art method by 5% in 5-shot segmentation.¹

1. Introduction

Humans have a remarkable ability to learn how to recognize novel objects after seeing only a handful of exemplars. On the other hand, deep learning based computer vision systems have made tremendous progress, but have largely depended on large-scale training sets. Also, deep networks mostly work with predefined classes and are incapable of generalizing to new ones. The field of few-shot learning studies the development of such learning ability in artificial learning systems, where only a few examples of the new category are available.