Landcover classification on sentinel-2 data with Prithvi, EfficientNet-Unet and OSM / CNES Landcover labels.
-
Updated
Apr 11, 2024 - Python
Landcover classification on sentinel-2 data with Prithvi, EfficientNet-Unet and OSM / CNES Landcover labels.
Fine-tuning Geospatial Foundation Models (Prithvi, TerraMind) for building footprint segmentation from Sentinel-2 using TerraTorch — Algiers case study
Temporal crop analysis and multi-cropping detection using Prithvi EO 2.0 embeddings, Sentinel-2 imagery, and unsupervised learning.
Geo-MLOps: PEFT Benchmark for Geospatial Foundation Models | 75 experiments on Prithvi-100M — LoRA fails, Houlsby dominates, modality > method
Reliability-aware biodiversity data integration using BBS, eBird, NASA Earthdata HLS/Sentinel-2, and Prithvi/TerraMind/Clay for trustworthy ecological inference.
NYC Hurricane Ida pluvial-flood pattern detector. Fine-tune of NASA-IBM Prithvi-EO 2.0 (300M params). Apache-2.0.
Geospatial foundation models, side by side — run SAM, Grounding DINO, Prithvi-EO & Clay on satellite/aerial imagery and score them against ground truth. Inference-only, runs locally.
21st place (of 116) solution for Solafune’s Construction Cost Prediction competition: predicting construction cost per m² in Japan and the Philippines from economic and satellite data. AutoGluon on the tabular data beat a Prithvi EO 2.0 embedding + XGBoost/LightGBM/CatBoost pipeline.
To associate your repository with the prithvi topic, visit your repo's landing page and select "manage topics."