Breaking the Box: Enhancing Remote Sensing Image Segmentation with Freehand Sketches
This work addresses the problem of making remote sensing image segmentation more intuitive and effective for users, though it is incremental by building on existing interactive segmentation approaches.
The paper tackles zero-shot interactive segmentation for remote sensing imagery by proposing a sketch-based prompting method, a new dataset, and a model, resulting in significant improvements in segmentation accuracy and robustness over state-of-the-art methods like SAM.
This work advances zero-shot interactive segmentation for remote sensing imagery through three key contributions. First, we propose a novel sketch-based prompting method, enabling users to intuitively outline objects, surpassing traditional point or box prompts. Second, we introduce LTL-Sensing, the first dataset pairing human sketches with remote sensing imagery, setting a benchmark for future research. Third, we present LTL-Net, a model featuring a multi-input prompting transport module tailored for freehand sketches. Extensive experiments show our approach significantly improves segmentation accuracy and robustness over state-of-the-art methods like SAM, fostering more intuitive human-AI collaboration in remote sensing analysis and enhancing its applications.