MapSR: Prompt-Driven Land Cover Map Super-Resolution via Vision Foundation Models

2026-04-16 · Source: Takara TLDR - Daily AI Papers · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Computer Vision, Geospatial AI · Depth: Advanced, medium

Summary

MapSR is a prompt-driven framework designed for super-resolution of land-cover maps, addressing the high cost of dense high-resolution (HR) annotations. Published on April 16, 2026, this method enhances coarse low-resolution (LR) land-cover products into HR maps at the resolution of input imagery. Unlike existing weakly supervised methods that retrain dense predictors with substantial computational cost, MapSR decouples supervision from model training. It extracts class prompts from frozen vision foundation model features using a lightweight linear probe, then performs HR mapping via training-free metric inference and graph-based prediction refinement. Experiments on the Chesapeake Bay dataset demonstrate MapSR achieves 59.64% mIoU without any HR labels, competing with strong weakly supervised baselines and outperforming fully supervised ones. Crucially, MapSR reduces trainable parameters by four orders of magnitude and shortens training time from hours to minutes, enabling scalable HR mapping under limited annotation and compute budgets.

Key takeaway

For research scientists developing land-cover mapping solutions, MapSR offers a compelling approach to achieve high-resolution outputs with significantly reduced annotation and computational demands. You should consider integrating prompt-driven frameworks with frozen vision foundation models to drastically cut training time and parameter counts, enabling more scalable and cost-effective mapping projects, especially when HR labels are scarce.

Key insights

MapSR enables high-resolution land-cover mapping with minimal annotation and compute by decoupling supervision from training.

Principles

• Decouple supervision from model training.
• Utilize frozen foundation model features.
• Refine predictions with graph-based propagation.

Method

MapSR extracts class prompts from frozen vision foundation model features via a linear probe, then performs HR mapping using training-free metric inference and graph-based spatial refinement.

In practice

• Apply linear probes to extract class prompts.
• Use cosine-similarity matching for HR predictions.
• Implement graph-based propagation for spatial refinement.

Topics

• Land Cover Mapping
• Map Super-Resolution
• Vision Foundation Models
• Prompt-Driven Learning
• Weakly Supervised Learning

Code references

• rikirikirikiriki/MapSR
• PeihaoWu/SOMA-1M
• mkhateri/Awesome-MRI-Super-Resolution

Best for: Research Scientist, AI Scientist, Machine Learning Engineer, Computer Vision Engineer

Related on AIssential

• MapSR: Prompt-Driven Land Cover Map Super-Resolution via Vision Foundation Models — MapSR uses prompt-driven vision foundation models for efficient, training-free land-cover super-resolution with minimal annotation.
• Scalable Deep Learning Framework for Global High-Resolution Land Use Reconstruction — AI4Land uses U-Net on HPC to reconstruct global 1 km land use from coarse data, improving climate model inputs.
• Beyond Visual Fidelity: Benchmarking Super-Resolution Models for Large-Scale Remote Sensing Imagery via Downstream Task Integration — Traditional SR metrics often fail to predict real-world performance in downstream remote sensing tasks.
• GMBFormer: An NDVI-Guided Global Memory Bank Transformer for Urban Green-Space Extraction from Ultra-High-Resolution Imagery — GMBFormer improves urban green-space extraction by decoupling NDVI and using a global memory bank for semantic reuse.
• Domain-Guided Prompting of the Segment Anything Model for Seismic Interpretation: The Role of Attributes, Visualization, and Hybrid Prompts — Zero-shot SAM adaptation for seismic interpretation is achieved by combining domain-guided visualization with hybrid prompting.
• Landsat-Sentinel-2 Algal Bloom Mapping Using Vision Transformers: Model Description, Implementation, and Examples — Vision Transformers enable robust, high-resolution algal bloom mapping from harmonized Landsat-Sentinel-2 imagery, outperforming traditional methods under challenging conditions.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Takara TLDR - Daily AI Papers.