# Enhanced 3D Object Detection via Diverse Feature Representations of 4D Radar Tensor

arXiv: https://arxiv.org/abs/2502.06114 · HTML: https://mlsanity.com/paper/2502.06114/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Detection and dense prediction distillation, one of 12 directions in Knowledge distillation.
- Builds on 21 of its 44 references that are in the map.
- First posted on arXiv 10 Feb 2025.

## Placed in

- Knowledge distillation / [Detection and dense prediction distillation](https://mlsanity.com/distill/detection-and-dense-prediction-distillation/)

## Builds on (most cited first, 21 of 21)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://mlsanity.com/paper/2010.04159/) (2020)
- [PointPillars: Fast Encoders for Object Detection from Point Clouds](https://mlsanity.com/paper/1812.05784/) (2018)
- [K-Radar: 4D Radar Object Detection for Autonomous Driving in Various Weather Conditions](https://mlsanity.com/paper/2206.08171/) (2022)
- [TJ4DRadSet: A 4D Radar Dataset for Autonomous Driving](https://mlsanity.com/paper/2204.13483/) (2022)
- [Scan Context++: Structural Place Recognition Robust to Rotation and Lateral Variations in Urban Environments](https://mlsanity.com/paper/2109.13494/) (2021)
- [Raw High-Definition Radar for Multi-Task Learning](https://mlsanity.com/paper/2112.10646/) (2021)
- [SMURF: Spatial Multi-Representation Fusion for 3D Object Detection with 4D Imaging Radar](https://mlsanity.com/paper/2307.10784/) (2023)
- [A Benchmark for Lidar Sensors in Fog: Is Detection Breaking Down?](https://mlsanity.com/paper/1912.03251/) (2019)
- [Robustness of Object Detectors in Degrading Weather Conditions](https://mlsanity.com/paper/2106.08795/) (2021)
- [RadarDistill: Boosting Radar-based Object Detection Performance via Knowledge Distillation from LiDAR Features](https://mlsanity.com/paper/2403.05061/) (2024)
- [CRKD: Enhanced Camera-Radar Object Detection with Cross-modality Knowledge Distillation](https://mlsanity.com/paper/2403.19104/) (2024)
- [T-FFTRadNet: Object Detection with Swin Vision Transformers from Raw ADC Radar Signals](https://mlsanity.com/paper/2303.16940/) (2023)
- [MVFAN: Multi-View Feature Assisted Network for 4D Radar Object Detection](https://mlsanity.com/paper/2310.16389/) (2023)
- [DPFT: Dual Perspective Fusion Transformer for Camera-Radar-based Object Detection](https://mlsanity.com/paper/2404.03015/) (2024)
- [Enhanced K-Radar: Optimal Density Reduction to Improve Detection Performance and Accessibility of 4D Radar Tensor-based Object Detection](https://mlsanity.com/paper/2303.06342/) (2023)
- [Sparse2Dense: Learning to Densify 3D Features for 3D Object Detection](https://mlsanity.com/paper/2211.13067/) (2022)
- [ADCNet: Learning from Raw Radar Data via Distillation](https://mlsanity.com/paper/2303.11420/) (2023)
- [MUFASA: Multi-View Fusion and Adaptation Network with Spatial Awareness for Radar Object Detection](https://mlsanity.com/paper/2408.00565/) (2024)

## Built on by (most cited first, 1 of 1)

- [On Learning Spatial Structure from Pre-Beamforming Per-Antenna Range-Doppler Radar Measurements](https://mlsanity.com/paper/2604.01921/) (2026)

## Abstract (arXiv)

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.