Long-tail Recognition via Compositional Knowledge Transfer

TL;DR

This paper proposes a training-free knowledge transfer method for long-tail recognition, leveraging class prototypes and cosine classifiers with an attention mechanism to improve rare class representations.

Contribution

It introduces a novel, training-free knowledge transfer approach that enhances rare class recognition by recomposing classifier features from common classes.

Findings

Significant performance improvements on rare classes

Maintains robust performance on common classes

Outperforms state-of-the-art models in long-tail recognition

Abstract

In this work, we introduce a novel strategy for long-tail recognition that addresses the tail classes' few-shot problem via training-free knowledge transfer. Our objective is to transfer knowledge acquired from information-rich common classes to semantically similar, and yet data-hungry, rare classes in order to obtain stronger tail class representations. We leverage the fact that class prototypes and learned cosine classifiers provide two different, complementary representations of class cluster centres in feature space, and use an attention mechanism to select and recompose learned classifier features from common classes to obtain higher quality rare class representations. Our knowledge transfer process is training free, reducing overfitting risks, and can afford continual extension of classifiers to new classes. Experiments show that our approach can achieve significant performance boosts on rare classes while maintaining robust common class performance, outperforming directly comparable state-of-the-art models.