dorsal/arxiv
View SchemaVibrantSR: Sub-Meter Canopy Height Models from Sentinel-2 Using Generative Flow Matching
| Authors | Kiarie Ndegwa, Andreas Gros, Tony Chang, David Diaz, Vincent A. Landau, Nathan E. Rutenbeck, Luke J. Zachmann, Guy Bayes, Scott Conway |
|---|---|
| Categories | |
| ArXiv ID | 2601.09866vv1 |
| URL | https://arxiv.org/abs/2601.09866 |
| License | http://creativecommons.org/licenses/by/4.0/ |
Abstract
We present VibrantSR (Vibrant Super-Resolution), a generative super-resolution framework for estimating 0.5 meter canopy height models (CHMs) from 10 meter Sentinel-2 imagery. Unlike approaches based on aerial imagery that are constrained by infrequent and irregular acquisition schedules, VibrantSR leverages globally available Sentinel-2 seasonal composites, enabling consistent monitoring at a seasonal-to-annual cadence. Evaluated across 22 EPA Level 3 eco-regions in the western United States using spatially disjoint validation splits, VibrantSR achieves a Mean Absolute Error of 4.39 meters for canopy heights >= 2 m, outperforming Meta (4.83 m), LANDFIRE (5.96 m), and ETH (7.05 m) satellite-based benchmarks. While aerial-based VibrantVS (2.71 m MAE) retains an accuracy advantage, VibrantSR enables operational forest monitoring and carbon accounting at continental scales without reliance on costly and temporally infrequent aerial acquisitions.
{
"annotation_id": "85e48f80-f2c5-4d4d-8057-e870b0579d44",
"date_created": "2026-02-17T05:53:23.751000Z",
"date_modified": "2026-02-17T05:53:23.751000Z",
"file_hash": "6f36a407ec26a5dc45af6ef7ad8871648080ce55af2a8548ee41661b548e90fc",
"private": false,
"record": {
"abstract": "We present VibrantSR (Vibrant Super-Resolution), a generative super-resolution framework for estimating 0.5 meter canopy height models (CHMs) from 10 meter Sentinel-2 imagery. Unlike approaches based on aerial imagery that are constrained by infrequent and irregular acquisition schedules, VibrantSR leverages globally available Sentinel-2 seasonal composites, enabling consistent monitoring at a seasonal-to-annual cadence. Evaluated across 22 EPA Level 3 eco-regions in the western United States using spatially disjoint validation splits, VibrantSR achieves a Mean Absolute Error of 4.39 meters for canopy heights \u003e= 2 m, outperforming Meta (4.83 m), LANDFIRE (5.96 m), and ETH (7.05 m) satellite-based benchmarks. While aerial-based VibrantVS (2.71 m MAE) retains an accuracy advantage, VibrantSR enables operational forest monitoring and carbon accounting at continental scales without reliance on costly and temporally infrequent aerial acquisitions.",
"arxiv_id": "2601.09866",
"authors": [
"Kiarie Ndegwa",
"Andreas Gros",
"Tony Chang",
"David Diaz",
"Vincent A. Landau",
"Nathan E. Rutenbeck",
"Luke J. Zachmann",
"Guy Bayes",
"Scott Conway"
],
"categories": [
"cs.CV",
"cs.LG"
],
"license": "http://creativecommons.org/licenses/by/4.0/",
"title": "VibrantSR: Sub-Meter Canopy Height Models from Sentinel-2 Using Generative Flow Matching",
"url": "https://arxiv.org/abs/2601.09866",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "07a77f44-1230-4dd0-8d96-f0b3661d13ce",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}