Explicit Attention-Enhanced Fusion for RGB-Thermal Perception Tasks

TL;DR

This paper introduces Explicit Attention-Enhanced Fusion (EAEF), a novel method for effectively combining RGB and thermal data in perception tasks, significantly improving performance across multiple benchmarks.

Contribution

The paper proposes a new fusion strategy that adaptively enhances feature extraction for different data availability scenarios, outperforming existing methods in perception tasks.

Findings

Outperforms state-of-the-art by 1.6% in mIoU on semantic segmentation

Achieves 3.1% improvement in MAE on salient object detection

Increases mAP by 2.3% on object detection

Abstract

Recently, RGB-Thermal based perception has shown significant advances. Thermal information provides useful clues when visual cameras suffer from poor lighting conditions, such as low light and fog. However, how to effectively fuse RGB images and thermal data remains an open challenge. Previous works involve naive fusion strategies such as merging them at the input, concatenating multi-modality features inside models, or applying attention to each data modality. These fusion strategies are straightforward yet insufficient. In this paper, we propose a novel fusion method named Explicit Attention-Enhanced Fusion (EAEF) that fully takes advantage of each type of data. Specifically, we consider the following cases: i) both RGB data and thermal data, ii) only one of the types of data, and iii) none of them generate discriminative features. EAEF uses one branch to enhance feature extraction for i) and iii) and the other branch to remedy insufficient representations for ii). The outputs of two branches are fused to form complementary features. As a result, the proposed fusion method outperforms state-of-the-art by 1.6\% in mIoU on semantic segmentation, 3.1\% in MAE on salient object detection, 2.3\% in mAP on object detection, and 8.1\% in MAE on crowd counting. The code is available at https://github.com/FreeformRobotics/EAEFNet.