Complementary Random Masking for RGB-Thermal Semantic Segmentation

TL;DR

This paper introduces a novel complementary random masking strategy and self-distillation loss for RGB-Thermal semantic segmentation, enhancing robustness and accuracy by encouraging the model to learn from both modalities even when partially available.

Contribution

It proposes a new masking and self-distillation approach that prevents over-reliance on a single modality in RGB-Thermal segmentation tasks.

Findings

Achieves state-of-the-art results on three benchmarks.

Improves robustness when one modality is partially missing.

Enhances the learning of complementary features from RGB and thermal data.

Abstract

RGB-thermal semantic segmentation is one potential solution to achieve reliable semantic scene understanding in adverse weather and lighting conditions. However, the previous studies mostly focus on designing a multi-modal fusion module without consideration of the nature of multi-modality inputs. Therefore, the networks easily become over-reliant on a single modality, making it difficult to learn complementary and meaningful representations for each modality. This paper proposes 1) a complementary random masking strategy of RGB-T images and 2) self-distillation loss between clean and masked input modalities. The proposed masking strategy prevents over-reliance on a single modality. It also improves the accuracy and robustness of the neural network by forcing the network to segment and classify objects even when one modality is partially available. Also, the proposed self-distillation loss encourages the network to extract complementary and meaningful representations from a single modality or complementary masked modalities. Based on the proposed method, we achieve state-of-the-art performance over three RGB-T semantic segmentation benchmarks. Our source code is available at https://github.com/UkcheolShin/CRM_RGBTSeg.