K-Shot Contrastive Learning of Visual Features with Multiple Instance Augmentations

TL;DR

This paper introduces K-Shot Contrastive Learning (KSCL), a novel method that leverages multiple augmentations and subspace modeling to improve visual feature discrimination and intra-instance variation handling, outperforming existing unsupervised methods.

Contribution

The paper proposes a new K-Shot contrastive learning framework that models intra-instance variations with subspaces and generalizes one-shot contrastive learning, trained end-to-end.

Findings

KSCL achieves superior performance over state-of-the-art unsupervised methods.

The method effectively models intra-instance variations through subspace configuration.

Experimental results validate the effectiveness of the proposed approach.

Abstract

In this paper, we propose the $K$-Shot Contrastive Learning (KSCL) of visual features by applying multiple augmentations to investigate the sample variations within individual instances. It aims to combine the advantages of inter-instance discrimination by learning discriminative features to distinguish between different instances, as well as intra-instance variations by matching queries against the variants of augmented samples over instances. Particularly, for each instance, it constructs an instance subspace to model the configuration of how the significant factors of variations in $K$-shot augmentations can be combined to form the variants of augmentations. Given a query, the most relevant variant of instances is then retrieved by projecting the query onto their subspaces to predict the positive instance class. This generalizes the existing contrastive learning that can be viewed as a special one-shot case. An eigenvalue decomposition is performed to configure instance subspaces, and the embedding network can be trained end-to-end through the differentiable subspace configuration. Experiment results demonstrate the proposed $K$-shot contrastive learning achieves superior performances to the state-of-the-art unsupervised methods.