iSAGE: A Human-in-the-Loop Framework for Remote Sensing Semantic Segmentation via Sparse Point Supervision
Researchers have developed iSAGE, a novel human-in-the-loop framework for remote sensing semantic segmentation. This system bypasses the need for costly pixel-level annotations by using expert clicks to target confident model errors, effectively amplifying the gradient at each click. Experiments demonstrate that iSAGE can achieve high accuracy with a minimal percentage of labeled pixels, outperforming existing methods on benchmark datasets. AI
IMPACT This framework could significantly reduce the cost and effort required for creating labeled datasets in remote sensing, accelerating the development and deployment of AI models in this domain.