Prototype Refinement Network for Few-Shot Segmentation

TL;DR

This paper introduces PRNet, a novel prototype refinement network for few-shot segmentation that improves prototype discrimination through adaptation and fusion, significantly enhancing performance on benchmark datasets.

Contribution

The paper proposes a prototype refinement method with adaptation and fusion techniques that do not require extra learnable parameters, advancing few-shot segmentation.

Findings

PRNet outperforms existing methods on COCO-20^i by 13.1% in 1-shot setting.

Prototype fusion effectively refines prototypes without additional learnable parameters.

Experiments demonstrate the superiority of PRNet on PASAL-5^i and COCO-20^i datasets.

Abstract

Few-shot segmentation targets to segment new classes with few annotated images provided. It is more challenging than traditional semantic segmentation tasks that segment known classes with abundant annotated images. In this paper, we propose a Prototype Refinement Network (PRNet) to attack the challenge of few-shot segmentation. It firstly learns to bidirectionally extract prototypes from both support and query images of the known classes. Furthermore, to extract representative prototypes of the new classes, we use adaptation and fusion for prototype refinement. The step of adaptation makes the model to learn new concepts which is directly implemented by retraining. Prototype fusion is firstly proposed which fuses support prototypes with query prototypes, incorporating the knowledge from both sides. It is effective in prototype refinement without importing extra learnable parameters. In this way, the prototypes become more discriminative in low-data regimes. Experiments on PASAL-$5^i$ and COCO-$20^i$ demonstrate the superiority of our method. Especially on COCO-$20^i$, PRNet significantly outperforms existing methods by a large margin of 13.1\% in 1-shot setting.