Enhanced 3D Object Detection via Diverse Feature Representations of 4D Radar Tensor
This work addresses efficiency and scalability issues for automotive perception systems using 4D Radar, representing an incremental improvement over existing methods.
The paper tackles the problem of high computational costs and limited scalability in 3D object detection using raw 4D Radar Tensor by proposing a multi-teacher knowledge distillation framework. It achieves improvements of 7.3% in AP_3D and 9.5% in AP_BEV over the baseline with sparse inputs, while reducing input data size by about 90 times.
Recent advances in automotive four-dimensional (4D) Radar have enabled access to raw 4D Radar Tensor (4DRT), offering richer spatial and Doppler information than conventional point clouds. While most existing methods rely on heavily pre-processed, sparse Radar data, recent attempts to leverage raw 4DRT face high computational costs and limited scalability. To address these limitations, we propose a novel three-dimensional (3D) object detection framework that maximizes the utility of 4DRT while preserving efficiency. Our method introduces a multi-teacher knowledge distillation (KD), where multiple teacher models are trained on point clouds derived from diverse 4DRT pre-processing techniques, each capturing complementary signal characteristics. These teacher representations are fused via a dedicated aggregation module and distilled into a lightweight student model that operates solely on a sparse Radar input. Experimental results on the K-Radar dataset demonstrate that our framework achieves improvements of 7.3% in AP_3D and 9.5% in AP_BEV over the baseline RTNH model when using extremely sparse inputs. Furthermore, it attains comparable performance to denser-input baselines while significantly reducing the input data size by about 90 times, confirming the scalability and efficiency of our approach.