GNN-JFL: Graph Neural Network for Video SAR Shadow Tracking With Joint Motion-Appearance Feature Learning

In this study, we address the challenges associated with video synthetic aperture radar (Video SAR) shadow tracking, a technique used for continuous monitoring of ground moving targets. Due to challenges such as changes in shadow appearance, low contrast between shadow and background, and scene occlusion in Video SAR, existing methods often encounter extensive matching errors in the data association process, resulting in unsatisfactory tracking performance. To overcome these issues, we propose a novel method, graph neural network – joint feature learning (GNN-JFL), which is based on joint motion-appearance feature extraction and graph neural data association. This method uses the detector as a flexible plugin and introduces two key improvements in the tracker section to enhance tracking accuracy. First, we introduce JFL to extract the complementary appearance and motion features from shadow shapes and positions, obtaining more robust feature representations to improve the tracking performance under intricate challenges. Second, by organically integrating multiobject tracking (MOT) problems and graph neural networks (GNNs), we propose a novel GNN-based shadow tracking architecture, which utilizes graph relationships to learn the associations between shadows for more accurate tracking predictions. Our method is validated using two measured datasets and demonstrates superior performance in terms of MOT accuracy (MOTA). It outperforms the suboptimal method by 4.2% and 3.6% in the two datasets. This research contributes to the advancement of continuous monitoring techniques employing Video SAR shadow tracking.