SAM-Aug: Leveraging SAM Priors for Few-Shot Parcel Segmentation in Satellite Time Series
Kai Hu, Yaozu Feng, Vladimir Lysenko, Ya Guo, and Huayi Wu

TL;DR
SAM-Aug introduces a novel framework that leverages the Segment Anything Model's priors to enhance few-shot land cover segmentation in satellite time series, significantly improving accuracy with minimal labeled data.
Contribution
The paper proposes SAM-Aug, a new annotation-efficient method that uses SAM-derived priors and a novel loss to improve few-shot remote sensing segmentation without additional annotations.
Findings
Outperforms state-of-the-art by +2.33 mIoU under 5% labels
Achieves up to 40.28% mIoU on the best seed
Demonstrates robustness across multiple random seeds
Abstract
Few-shot semantic segmentation of time-series remote sensing images remains a critical challenge, particularly in regions where labeled data is scarce or costly to obtain. While state-of-the-art models perform well under full supervision, their performance degrades significantly under limited labeling, limiting their real-world applicability. In this work, we propose SAM-Aug, a new annotation-efficient framework that leverages the geometry-aware segmentation capability of the Segment Anything Model (SAM) to improve few-shot land cover mapping. Our approach constructs cloud-free composite images from temporal sequences and applies SAM in a fully unsupervised manner to generate geometry-aware mask priors. These priors are then integrated into training through a proposed loss function called RegionSmoothLoss, which enforces prediction consistency within each SAM-derived region across…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsRemote-Sensing Image Classification · Remote Sensing in Agriculture · Automated Road and Building Extraction
