# UniDistill: A Universal Cross-Modality Knowledge Distillation Framework for 3D Object Detection in Bird's-Eye View

Source: [UniDistill: A Universal Cross-Modality Knowledge Distillation Framework for 3D Object Detection in Bird's-Eye View](<https://www.sota2.com/research/paper/68fa6ede1e209131cea427fa>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-04T04:28:48.367Z

Authors: Shengchao Zhou, Weizhou Liu, Chen Hu, Shuchang Zhou, Chao Ma

Published: 2023-03-27T10:50:58.000Z

## Abstract

In the field of 3D object detection for autonomous driving, the sensor portfolio including multi-modality and single-modality is diverse and complex. Since the multi-modal methods have system complexity while the accuracy of single-modal ones is relatively low, how to make a tradeoff between them is difficult. In this work, we propose a universal cross-modality knowledge distillation framework (UniDistill) to improve the performance of single-modality detectors. Specifically, during training, UniDistill projects the features of both the teacher and the student detector into Bird's-Eye-View (BEV), which is a friendly representation for different modalities. Then, three distillation losses are calculated to sparsely align the foreground features, helping the student learn from the teacher without introducing additional cost during inference. Taking advantage of the similar detection paradigm of different detectors in BEV, UniDistill easily supports LiDAR-to-camera, camera-to-LiDAR, fusion-to-LiDAR and fusion-to-camera distillation paths. Furthermore, the three distillation losses can filter the effect of misaligned background information and balance between objects of different sizes, improving the distillation effectiveness. Extensive experiments on nuScenes demonstrate that UniDistill effectively improves the mAP and NDS of student detectors by 2.0%~3.2%.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [3D Object Detection on nuScenes (test)](<https://www.sota2.com/research/sota/3d-object-detection-on-nuscenes-test.md>): 3D Object Detection; nuScenes (test); mAP: 65.4.

- [3D Object Detection on nuScenes (val)](<https://www.sota2.com/research/sota/3d-object-detection-on-nuscenes-val.md>): 3D Object Detection; nuScenes (val); NDS: 37.8.

- [3D Object Detection on Waymo Open Dataset (val)](<https://www.sota2.com/research/sota/3d-object-detection-on-waymo-open-dataset-val.md>): 3D Object Detection; Waymo Open Dataset (val); 3D APH Vehicle L2: not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2303.15083>)

[PDF](<https://arxiv.org/pdf/2303.15083.pdf>)

[Implementation](<https://github.com/megvii-research/CVPR2023-UniDistill>)