# Teacher-Student Knowledge Distillation for Radar Perception on Embedded Accelerators

arXiv: https://arxiv.org/abs/2303.07586 · HTML: https://mlsanity.com/paper/2303.07586/

## 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 9 of its 19 references that are in the map.
- First posted on arXiv 14 Mar 2023.

## Placed in

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

## Builds on (most cited first, 9 of 9)

- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Knowledge Distillation and Student-Teacher Learning for Visual Intelligence: A Review and New Outlooks](https://mlsanity.com/paper/2004.05937/) (2020)
- [An Intriguing Failing of Convolutional Neural Networks and the CoordConv Solution](https://mlsanity.com/paper/1807.03247/) (2018)
- [3D Object Detection for Autonomous Driving: A Survey](https://mlsanity.com/paper/2106.10823/) (2021)
- [Seeing Around Street Corners: Non-Line-of-Sight Detection and Tracking In-the-Wild Using Doppler Radar](https://mlsanity.com/paper/1912.06613/) (2019)
- [Object Detection and 3D Estimation via an FMCW Radar Using a Fully Convolutional Network](https://mlsanity.com/paper/1902.05394/) (2019)
- [DeepHybrid: Deep Learning on Automotive Radar Spectra and Reflections for Object Classification](https://mlsanity.com/paper/2202.08519/) (2022)
- [Self-Supervised Velocity Estimation for Automotive Radar Object Detection Networks](https://mlsanity.com/paper/2207.03146/) (2022)
- [Fusion of Deep Neural Networks for Activity Recognition: A Regular Vine Copula Based Approach](https://mlsanity.com/paper/1905.02703/) (2019)

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

- [REOcc: Camera-Radar Fusion with Radar Feature Enrichment for 3D Occupancy Prediction](https://mlsanity.com/paper/2511.06666/) (2025)

## Abstract (arXiv)

Many radar signal processing methodologies are being developed for critical road safety perception tasks. Unfortunately, these signal processing algorithms are often poorly suited to run on embedded hardware accelerators used in automobiles. Conversely, end-to-end machine learning (ML) approaches better exploit the performance gains brought by specialized accelerators. In this paper, we propose a teacher-student knowledge distillation approach for low-level radar perception tasks. We utilize a hybrid model for stationary object detection as a teacher to train an end-to-end ML student model. The student can efficiently harness embedded compute for real-time deployment. We demonstrate that the proposed student model runs at speeds 100x faster than the teacher model.