UAV-to-map registration
The frames come with an approximate GPS position but no camera pose. Dense matching against satellite imagery gives each one a homography to the ground, so every pixel has a latitude and longitude, and every map entity can be projected into the image.
One row per frame
The unified frame table, frames.csv, records each step. Read it left to right: what the drone gave us, how it was registered, the footprint that came out, what follows from the footprint, and what SAM3 found in the image.
aGNSS prior
The image and the position the drone logged. Altitude references differ by source.
image_pathwidthheightdatetimegnss_latgnss_lonalt_malt_ref
bDense matching
RoMa matches the frame to satellite imagery around the prior; RANSAC fits a homography.
reg_methodreg_scorereg_status
cFootprint
The four image corners on the ground: tl is pixel (0,0), then clockwise.
tl_lattl_lontr_…br_…bl_…center_latcenter_lon
dDerived geometry
Heading of the image's up direction, edge lengths, area and ground resolution.
heading_degfootprint_w_mfootprint_h_mfootprint_area_m2gsd_m
eDetections
SAM3 buildings and cars, kept only if there are at least four buildings.
n_buildingn_carsam3_jsoncamerafx_px
Drape a frame over the ground
Hand-checked registrations from each source, chosen to cover different scenes. The UAV image is warped onto its four registered corners: fade it against the satellite basemap to check the fit, switch to OpenStreetMap to see the map prior it produces, and compare the GNSS position with where the image landed.
How good is it?
Six annotators georeferenced a stratified sample of 1,440 benchmark frames by hand in QGIS with ground control points. Comparing the manual and automatic transformations on a 64 × 64 grid per image gives a median error of 1.9 m.
Grid points within each error threshold
automatic vs manual registration, 1,440 images
| Median | 25th pct | 75th pct | 95th pct |
|---|---|---|---|
| 1.9 m | 1.3 m | 2.6 m | 5.4 m |
A few metres is enough to place the map window and to estimate the heading. It is not enough on its own for instance labels in dense neighbourhoods, where residual error, roof–footprint parallax or relief displacement can move a projected footprint onto the building next door. That is why the next stage never trusts a projected polygon by itself.
Before this check, three coauthors replayed every flight with its projected map overlay and discarded frames with visible drift, a wrong homography or a misaligned basemap tile, on top of the automatic checks: too few inliers, or a non-convex, degenerate, over-skewed or wrongly sized footprint.
Registration across the pool
Two registration pipelines produce the corners. USGS frames use heading, altitude and position priors from the survey metadata; AirLock, VETRA and UAV-VisLoc use a rotation sweep against satellite tiles. Inlier counts are comparable within a pipeline, not across them.
Match inliers
RANSAC inliers of the winning RoMa match, log scale
GNSS position to footprint centre
distance in metres between the logged GNSS position and the registered footprint centre, log scale
From candidate frames to the pool
all four sources