Spectral-Spatial Contrastive Learning Framework for Regression on Hyperspectral Data

TL;DR

This paper introduces a spectral-spatial contrastive learning framework tailored for regression tasks on hyperspectral data, enhancing various backbone models and demonstrating significant performance improvements on synthetic and real datasets.

Contribution

It presents a novel, model-agnostic contrastive learning framework with specialized hyperspectral data augmentations for regression tasks.

Findings

Significant performance improvements on synthetic datasets

Enhanced results on real hyperspectral datasets

Effective with multiple backbone architectures

Abstract

Contrastive learning has demonstrated great success in representation learning, especially for image classification tasks. However, there is still a shortage in studies targeting regression tasks, and more specifically applications on hyperspectral data. In this paper, we propose a spectral-spatial contrastive learning framework for regression tasks for hyperspectral data, in a model-agnostic design allowing to enhance backbones such as 3D convolutional and transformer-based networks. Moreover, we provide a collection of transformations relevant for augmenting hyperspectral data. Experiments on synthetic and real datasets show that the proposed framework and transformations significantly improve the performance of all studied backbone models.