Self-Promoted Prototype Refinement for Few-Shot Class-Incremental Learning

TL;DR

This paper introduces a novel prototype refinement scheme for few-shot class-incremental learning, improving recognition of new classes with limited samples while maintaining old class knowledge.

Contribution

It proposes a self-promoted prototype refinement mechanism with a dynamic relation projection module, enhancing prototype expression and incremental learning performance.

Findings

Outperforms state-of-the-art methods by 11-17% on benchmark datasets.

Effective in handling few-shot class-incremental learning challenges.

Demonstrates significant improvements in incremental learning accuracy.

Abstract

Few-shot class-incremental learning is to recognize the new classes given few samples and not forget the old classes. It is a challenging task since representation optimization and prototype reorganization can only be achieved under little supervision. To address this problem, we propose a novel incremental prototype learning scheme. Our scheme consists of a random episode selection strategy that adapts the feature representation to various generated incremental episodes to enhance the corresponding extensibility, and a self-promoted prototype refinement mechanism which strengthens the expression ability of the new classes by explicitly considering the dependencies among different classes. Particularly, a dynamic relation projection module is proposed to calculate the relation matrix in a shared embedding space and leverage it as the factor for bootstrapping the update of prototypes. Extensive experiments on three benchmark datasets demonstrate the above-par incremental performance, outperforming state-of-the-art methods by a margin of 13%, 17% and 11%, respectively.