SSeg: Active Sparse Point-Label Augmentation for Semantic Segmentation

ArXi:2510.10163v2 Announce Type: replace Semantic segmentation is essential for automating remote sensing analysis in fields like ecology. However, fine-grained analysis of complex aerial or underwater imagery remains an open challenge, even for state-of-the-art models. Progress is frequently hindered by the high cost of obtaining the dense, expert-annotated labels required for model supervision. While sparse point-labels are easier to obtain, they