RGB-Sonar Tracking Benchmark and Spatial Cross-Attention Transformer Tracker

Underwater camera and sonar are naturally complementary in the underwater environment. Combining the information from two modalities will promote better observation of underwater targets. However, this problem has received little attention in previous research. Therefore, this paper introduces a new and challenging RGB-Sonar (RGB-S) tracking task and investigates how to achieve efficient tracking of an underwater target through the interaction of the RGB and sonar modalities. Specifically, we first propose an RGBS50 benchmark dataset containing 50 sequences and more than 87,000 high-quality annotated bounding boxes. Experimental results show that the RGBS50 benchmark poses significant challenges to the currently popular SOT trackers. Second, we propose two RGB-S trackers, which are called SCANet and SCANet-Refine. They include a spatial cross-attention module (SCAM) consisting of a novel spatial cross-attention layer, an attention refinement module, and two independent global integration modules. The spatial cross-attention is used to overcome the problem of spatial misalignment between RGB and sonar images. Third, we propose a SOT data-based RGB-S simulation training method (SRST) to overcome the lack of RGB-S training datasets. It converts RGB images into sonar-like saliency images to construct pseudo-data pairs, enabling the model to learn the semantic structure of RGB-S data. Comprehensive experiments show that the proposed spatial cross-attention effectively achieves the interaction between RGB and sonar modalities, and that SCANet and SCANet-Refine achieves state-of-the-art performance on the proposed benchmark. The code is available at https://github.com/LiYunfengLYF/RGBS50.