# Generative detection of ghost roads with Tessera embeddings

*2026-08-01 — idea*


Roads are expanding faster than at any point in human history, and many of them
don't appear on maps. The [ghost roads](https://doi.org/10.1038/s41586-024-07303-5)
study in 2024 used ~7,000 hours of volunteer effort to hand-map roads across the
tropical Asia-Pacific, finding 3--6.6 times more roads than exist in current global
road datasets.

Road building almost always preceded local forest loss, making unmapped roads
one of the best early warning signals we have for deforestation. Mapping them
quickly and globally is therefore a conservation priority and not just a
"cartographic tidyup"\!

The current SOTA detects roads directly from Sentinel imagery with
CNN segmentation models, such as [monitoring road development in the Congo Basin](https://doi.org/10.1016/j.rse.2024.114380),
[mapping roads in the Brazilian Amazon](https://doi.org/10.3390/rs14153625), and
the ghost roads team's own [remote road mapping with AI](https://doi.org/10.3390/rs16050839).
The 10m resolution of Sentinel does a surprisingly good job, esp. for new roads
where there's an initial canopy opening. It's also free and global.

This project would investigate whether the [TESSERA](https://anil.recoil.org/projects/tessera) [foundation model](https://anil.recoil.org/papers/2026-tessera-v2) can be better/faster than raw Sentinel imagery. Tessera compresses a
full year of Sentinel-1 and Sentinel-2 observations into a 128-dimensional
embedding for every 10m pixel on the planet (easily accessible via
[GeoTessera Python library released for geospatial embeddings](https://anil.recoil.org/notes/geotessera-python)), so a road classifier also gets the radar signal through
cloud and a year of temporal dynamics rather than a single optical scene.

That temporal signal might also allow us to separate paved from unpaved
surfaces, and to spot informal logging tracks that only appear seasonally. Road
detection is already one of the downstream benchmarks in the Tessera v2
evaluation suite (via the [CityRep](https://arxiv.org/abs/2605.26036) urban
representation benchmark), so there is a baseline to start from.

An interesting twist with roads is that they are not independent
pixels. Almost every road connects to another road! A per-pixel classifier
throws that structural prior away, but a generative approach could
exploit it by starting from a known road network (e.g. OpenStreetMap) and
tracing outwards into the unknown. Exactly which ML technique fits best is an
open question the project would settle, but candidates include:

- Connectivity-aware segmentation by adding topology-preserving losses to a
  conventional segmentation head over embedding patches so that predicted roads
  form connected networks rather than a speckle.
- Iterative tracing in the style of [RoadTracer](https://arxiv.org/abs/1802.03680),
  where an agent walks outwards from known road seeds and decides at each step
  whether the road continues through embedding space.
- Generative graph completion, treating the mapped network as a partial
  graph and training a model to propose the missing edges, sampling multiple
  plausible completions to get uncertainty over where ghost roads might lurk.
- Temporal change detection, differencing yearly Tessera embeddings to
  catch new canopy openings as roads are actively constructed.

Validation can be done since the ghost roads team's hand-digitised dataset gives us
high-quality labels for where roads actually are across the Asia-Pacific tropics, and
beyond what OpenStreetMap coverage would suggest for those years. We have Tessera
embeddings from 2017-2025.

## See also

[Andres Zuñiga-Gonzalez](https://ancazugo.github.io/) has done prior work on road mapping with OSM labels over Tessera
embeddings; see his weeknotes on [detecting new roads](https://ancazugo.github.io/posts/2025-11-23-weekly-notes.html)
and [categorising OSM road features](https://ancazugo.github.io/posts/2026-02-15-weekly-notes.html),
and his [osm-rasterizer](https://github.com/ancazugo/osm-rasterizer) tool for
generating training rasters from OSM features.

The [Global habitat maps from Tessera embeddings](https://anil.recoil.org/ideas/tessera-habitat-maps) work covers the broader classifier-on-embeddings
methods we're developing as part of the wider [TESSERA](https://anil.recoil.org/projects/tessera) programme.
Status: Available
Level: MPhil
Year: 2026
Project: TESSERA, a pixelwise geospatial foundation model
Supervisors: Anil Madhavapeddy, Sadiq Jaffer, Jayden Engert, David Coomes

## Related

- [Global habitat maps from Tessera embeddings](https://anil.recoil.org/ideas/tessera-habitat-maps) (idea, 2026-08-01)
- [TESSERA v2: Scaling Pixel-wise Earth Foundation Models](https://anil.recoil.org/papers/2026-tessera-v2) (paper, 2026-07-01)
- [GeoTessera Python library released for geospatial embeddings](https://anil.recoil.org/notes/geotessera-python) (note, 2025-08-31)
- [TESSERA, a pixelwise geospatial foundation model](https://anil.recoil.org/projects/tessera) (project, 2025-01-01)

---
Canonical: https://anil.recoil.org/ideas/tessera-ghost-roads
Type: idea
Tags: tessera, biodiversity, conservation, ai, spatial