UniBEV: Multi-modal 3D Object Detection with Uniform BEV Encoders for Robustness Against Missing Sensor Modalities

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

Abstract

Multi-sensor object detection is an active research topic in automated driving, but the robustness of such detection models against missing sensor input (modality missing), e.g., due to a sudden sensor failure, is a critical problem which remains under-studied. In this work, we propose UniBEV, an end-to-end multi-modal 3D object detection framework designed for robustness against missing modalities: UniBEV can operate on LiDAR plus camera input, but also on LiDAR-only or camera-only input without retraining. To facilitate its detector head to handle different input combinations, UniBEV aims to create well-aligned Bird’s Eye View (BEV) feature maps from each available modality. Unlike prior BEV-based multi-modal detection methods, all sensor modalities follow a uniform approach to resample features from the original sensor coordinate systems to the BEV features. We furthermore investigate the robustness of various fusion strategies w.r.t. missing modalities: the commonly used feature concatenation, but also channel-wise averaging, and a generalization to weighted averaging termed Channel Normalized Weights. To validate its effectiveness, we compare UniBEV to state-of-the-art BEVFusion and MetaBEV on nuScenes over all sensor input combinations. In this setting, UniBEV achieves better performance than these baselines for all input combinations. An ablation study shows the robustness benefits of fusing by weighted averaging over regular concatenation, and of sharing queries between the BEV encoders of each modality. Our code is available at https://github.com/tudelft-iv/UniBEV.
Original languageEnglish
Title of host publication35th IEEE Intelligent Vehicles Symposium, IV 2024
PublisherIEEE
Pages2776-2783
Number of pages8
ISBN (Electronic)979-8-3503-4881-1
DOIs
Publication statusPublished - 2024

Publication series

NameIEEE Intelligent Vehicles Symposium, Proceedings
PublisherIEEE
ISSN (Print)1931-0587
ISSN (Electronic)2642-7214

Bibliographical note

Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

  • solid modeling
  • three-dimensional displays
  • laser radar
  • codes
  • object detection
  • detectors
  • feature extraction

Fingerprint

Dive into the research topics of 'UniBEV: Multi-modal 3D Object Detection with Uniform BEV Encoders for Robustness Against Missing Sensor Modalities'. Together they form a unique fingerprint.

Cite this