Enhancing Radar Perception through Multi-Task Learning: Refining Data for Sensor Fusion Applications
This work introduces a learning-based approach to accurately estimate the height of radar points, enabling refined radar data for downstream perception tasks such as object detection and depth estimation.