Global Rectification and Decoupled Registration for Few-Shot Segmentation in Remote Sensing Imagery

Few-shot segmentation (FSS), which aims to determine specific objects in the query image given only a handful of densely labeled samples, has received extensive academic attention in recent years. However, most existing FSS methods are designed for natural images, and few works have been done to investigate more realistic and challenging applications, e.g., remote sensing image understanding. In such a setup, the complex nature of the raw images would undoubtedly further increase the difficulty of the segmentation task. To couple with potential inference failures, we propose a novel and powerful remote sensing FSS framework with global rectification (GR) and decoupled registration (DR), termed R2Net. Specifically, a series of dynamically updated global prototypes are utilized to provide auxiliary nontarget segmentation cues and to prevent inaccurate prototype activation resulting from the variability between query-support image pairs. The foreground (FG) and background information flows are then decoupled for more targeted and tailored object localization, avoiding unnecessary confusion from information redundancy. Furthermore, we impose additional constraints to promote interclass separability and intraclass compactness. Extensive experiments on the standard benchmark iSAID- $5^{i}$ demonstrate the superiority of the proposed R2Net over state-of-the-art FSS models. The code is available at https://github.com/chunbolang/R2Net.