RT-X Net: RGB-Thermal cross attention network for Low-Light Image Enhancement

TL;DR

RT-X Net is a novel cross-attention network that fuses RGB and thermal images to significantly improve low-light image enhancement in nighttime conditions, supported by a new dataset and extensive evaluations.

Contribution

The paper introduces RT-X Net, a cross-attention based fusion model for RGB and thermal images, and provides a new dataset for low-light image enhancement research.

Findings

RT-X Net outperforms existing methods on LLVIP and V-TIEE datasets.

The cross-attention mechanism effectively integrates multi-modal information.

The V-TIEE dataset supports future research in thermal and visible image enhancement.

Abstract

In nighttime conditions, high noise levels and bright illumination sources degrade image quality, making low-light image enhancement challenging. Thermal images provide complementary information, offering richer textures and structural details. We propose RT-X Net, a cross-attention network that fuses RGB and thermal images for nighttime image enhancement. We leverage self-attention networks for feature extraction and a cross-attention mechanism for fusion to effectively integrate information from both modalities. To support research in this domain, we introduce the Visible-Thermal Image Enhancement Evaluation (V-TIEE) dataset, comprising 50 co-located visible and thermal images captured under diverse nighttime conditions. Extensive evaluations on the publicly available LLVIP dataset and our V-TIEE dataset demonstrate that RT-X Net outperforms state-of-the-art methods in low-light image enhancement. The code and the V-TIEE can be found here https://github.com/jhakrraman/rt-xnet.