Dual Path Structural Contrastive Embeddings for Learning Novel Objects

TL;DR

This paper introduces a dual path contrastive embedding method that enhances feature representations for few-shot learning by balancing intra-class similarity and inter-class discrimination, leading to improved performance on standard benchmarks.

Contribution

It proposes a novel dual path feature learning scheme that decouples feature learning from classification, combining structural similarity with contrastive learning for better generalization.

Findings

Achieves promising results on three benchmarks.

Effective in both standard and generalized few-shot tasks.

Works well in inductive and transductive settings.

Abstract

Learning novel classes from a very few labeled samples has attracted increasing attention in machine learning areas. Recent research on either meta-learning based or transfer-learning based paradigm demonstrates that gaining information on a good feature space can be an effective solution to achieve favorable performance on few-shot tasks. In this paper, we propose a simple but effective paradigm that decouples the tasks of learning feature representations and classifiers and only learns the feature embedding architecture from base classes via the typical transfer-learning training strategy. To maintain both the generalization ability across base and novel classes and discrimination ability within each class, we propose a dual path feature learning scheme that effectively combines structural similarity with contrastive feature construction. In this way, both inner-class alignment and inter-class uniformity can be well balanced, and result in improved performance. Experiments on three popular benchmarks show that when incorporated with a simple prototype based classifier, our method can still achieve promising results for both standard and generalized few-shot problems in either an inductive or transductive inference setting.