Semantic classification of automotive radar detections using the RadarScenes dataset. The pipeline extracts point-level and track-level features from raw radar sequences and benchmarks several supervised classifiers (XGBoost, Random Forest, Logistic Regression, Gaussian NB) with strict sequence-level splitting to prevent data leakage.
radar machine-learning-algorithms regression supervised-learning xgboost classification logistic-regression automotive multiclass-classification supervised-machine-learning multiclass radar-data radarscenes semantic-prediction
-
Updated
Jun 16, 2026 - HTML