RaMOD: Moving-Object Detection from Raw mmWave Radar Cubes under Ego Motion
Overview
This research presents RaMOD, a real-time framework for detecting moving objects and estimating velocities from raw range-azimuth-Doppler (RAD) cubes acquired by vehicle-mounted single-chip mmWave radars. The method explicitly estimates frame-wise ego motion to predict and suppress Doppler responses from world-static structures, then temporally encodes the suppressed data into multi-scale features for object detection. For each detected object, it predicts radial and forward velocities under both raw-Doppler and Doppler-unwrapped hypotheses along with Doppler-wrapping probabilities. Evaluated on the BRIDGE-Liv dataset, RaMOD achieves an 80.8% cube mAP at IoU 0.30 (15.7 percentage points above the strongest baseline), a velocity mean absolute error of 1.128 m/s, and 99.87% Doppler-wrapping accuracy.
Generated from attributed reports · Updated 5 hours ago
Event evidence and corrections
0 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.
Reported quantity · velocity MAE: 1.128 other · Basis not reported
“RA, and RD detection settings, achieving 80.8% cube mAP at IoU 0.30, 15.7 percentage points above the strongest RAD-cube baseline. Velocity MAE is 1.128 m/s, with 99.87% Doppler-wrapping accuracy.”
Exact source · revision 1Source owner not reported
Reported quantity · percentage points above baseline: 15.7 other · Basis not reported
“RA, and RD detection settings, achieving 80.8% cube mAP at IoU 0.30, 15.7 percentage points above the strongest RAD-cube baseline. Velocity MAE is 1.128 m/s, with 99.87% Doppler-wrapping accuracy.”
Exact source · revision 1Source owner not reported
Report timeline
Follow attributed reports and material updates.
- arXiv Robotics — research abstractsSignalRaMOD: Moving-Object Detection from Raw mmWave Radar Cubes under Ego Motion
This research presents RaMOD, a real-time framework for detecting moving objects and estimating velocities from raw range-azimuth-Doppler (RAD) cubes acquired by vehicle-mounted single-chip mmWave radars. The method explicitly estimates frame-wise ego motion to predict and suppress Doppler responses from world-static structures, then temporally encodes the suppressed data into multi-scale features for object detection. For each detected object, it predicts radial and forward velocities under both raw-Doppler and Doppler-unwrapped hypotheses along with Doppler-wrapping probabilities. Evaluated on the BRIDGE-Liv dataset, RaMOD achieves an 80.8% cube mAP at IoU 0.30 (15.7 percentage points above the strongest baseline), a velocity mean absolute error of 1.128 m/s, and 99.87% Doppler-wrapping accuracy.
Event coverage history
There is not enough continuous observation data to show a trend.
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).