jaychempan/EarthSynth
Official Code for “EarthSynth: Generating Informative Earth Observation with Diffusion Models”
This project helps remote sensing analysts and researchers overcome the challenge of limited labeled satellite imagery for training their interpretation models. It takes descriptions of ground features and generates synthetic, high-quality satellite images and corresponding mask data. This allows users to create diverse training datasets, improving the performance of tasks like identifying roads, vehicles, or specific land covers in real-world satellite imagery.
Use this if you need to generate realistic, labeled satellite imagery for training or testing remote sensing models, especially when real-world labeled data is scarce.
Not ideal if you primarily need to analyze existing satellite data without generating new synthetic images.
Stars
54
Forks
1
Language
Python
License
MIT
Category
Last pushed
Oct 30, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/jaychempan/EarthSynth"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
cswry/SeeSR
[CVPR2024] SeeSR: Towards Semantics-Aware Real-World Image Super-Resolution
JJLibra/SALAD-Pan
🤗 Official implementation for "SALAD-Pan: Sensor-Agnostic Latent Adaptive Diffusion for...
open-mmlab/mmgeneration
MMGeneration is a powerful toolkit for generative models, based on PyTorch and MMCV.
Janspiry/Image-Super-Resolution-via-Iterative-Refinement
Unofficial implementation of Image Super-Resolution via Iterative Refinement by Pytorch
hanjq17/Spectrum
[CVPR 2026] Adaptive Spectral Feature Forecasting for Diffusion Sampling Acceleration