Report Cards

Use test range trajectories to see where a detection and tracking stack holds or fails. Group 1, Group 2, Group 3. Completely offline.

Offline Benchmarking

Example report card · Measured

UAS Detection Model · RGB Day Intercepts

Model
vigil-association-transformer-v1
Release
rgb-day-intercepts-v1
Evaluation split
held-out
Spectrum
RGB

Detection evaluation

12,480 scored frames

mAP@[.50:.95]
0.742
AP50
0.942
Calibration ECE
0.0410
P95 latency
18.00 ms

Single-object tracking

8,112 scored frames

Success
0.874
Precision
0.912
Normalized precision
0.889
State accuracy
0.931

Offline Benchmarking completed. This report makes no System Qualification decision.

Completed Aug 11, 2026, 2:14 PM

Neural Sensing Engine™

Replay your unique sensor stack against our corpus.

We remap the capture onto the payload you ship and run the stack you will actually field. Completely offline.

Partner with usGet Data

Licensed models

UAV Detection Model

Licensed detection remapped to the payload you ship.

mAP@0.5
94.2%
Inference
120 fps
Model size
47 MB

UAV Tracking Model

Single-object tracking on the same remap. Holds the airframe across the trajectory.

MOTA
87.4%
Success
91.2%
P95 latency
18 ms