$K$K-Shot Contrastive Learning of Visual Features With Multiple Instance Augmentations

Haohang Xu, Hongkai Xiong, Guo-Jun Qi

Abstract: In this paper, we propose the <inline-formula><tex-math notation="LaTeX">$K$</tex-math></inline-formula> -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 <i>inter-instance discrimination</i> by learning discriminative features to distinguish between different instances, as well as <i>intra-instance variations</i> 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 <inline-formula><tex-math notation="LaTeX">$K$</tex-math></inline-formula> -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 <inline-formula><tex-math notation="LaTeX">$K$</tex-math></inline-formula> -shot contrastive learning achieves superior performances to the state-of-the-art unsupervised methods.