metadata
library_name: tte
pipeline_tag: feature-extraction
tags:
- location-encoder
- geospatial
- remote-sensing
- sentinel-2
- voronoi
license: mit
Tessellating the Earth (TTE) — location encoder
Maps a geographic coordinate (lat, lon) to a 512-dim embedding via a learnable
Spherical Voronoi partition of S² with global semantic tokens. ECCV 2026.
- Code: https://github.com/mvrl/TTE
- Project page: https://dcher95.github.io/TTE/
import torch
from tte import TTE # pip install from github.com/mvrl/TTE
model = TTE.from_pretrained("dcher95/TTE").eval()
coords = torch.tensor([[37.77, -122.42], # (lat, lon) in degrees
[-3.12, 60.02]])
emb = model.encode(coords) # (N, 512), L2-normalized
Image backbone used during training (not needed here): frozen SSL4EO-S12 MAE ViT-L/16.