Views
No views yet
(lat, lon) to a learned embedding via a learnable
Spherical Voronoi partition of S² with global semantic tokens. ECCV 2026.1import torch
2from tte import TTE
3
4model = TTE.from_pretrained("MVRL/TTE").eval()
5coords = torch.tensor([[37.77, -122.42],
6 [-3.12, 60.02]])
7emb = model.encode(coords)1@inproceedings{cher2026tte,
2 title = {Tessellating the Earth: Learnable Spherical Voronoi Partitions for Location Encoding},
3 author = {Cher, Daniel and Iqbal, Hamza and Xing, Eric and Wei, Brian and Jacobs, Nathan},
4 booktitle = {European Conference on Computer Vision (ECCV)},
5 year = {2026}
6}