Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

LPCF-YOLO: A YOLO-Based Lightweight Algorithm for Pedestrian Anomaly Detection with Parallel Cross-Fusion.

Sensors (Basel, Switzerland)·2025
Same author

A <i>Schistosoma japonicum</i> MicroRNA Exerts Antitumor Effects Through Inhibition of Both Cell Migration and Angiogenesis by Targeting PGAM1.

Frontiers in oncology·2021
Same author

Prediction of skin disease using a new cytological taxonomy based on cytology and pathology with deep residual learning method.

Scientific reports·2021
Same author

Single cell imaging reveals cisplatin regulating interactions between transcription (co)factors and DNA.

Chemical science·2021
Same author

Succession of marine bacteria in response to Ulva prolifera-derived dissolved organic matter.

Environment international·2021
Same author

Increased TIGIT expressing NK cells with dysfunctional phenotype in AML patients correlated with poor prognosis.

Cancer immunology, immunotherapy : CII·2021

Related Experiment Video

Updated: Jun 13, 2026

Long-term Video Tracking of Cohoused Aquatic Animals: A Case Study of the Daily Locomotor Activity of the Norway Lobster (Nephrops norvegicus)
05:57

Long-term Video Tracking of Cohoused Aquatic Animals: A Case Study of the Daily Locomotor Activity of the Norway Lobster (Nephrops norvegicus)

Published on: April 8, 2019

SpecBEV: An End-to-End BEV 3D Object Detection Algorithm Based on Frequency-Domain Analysis and Geometric Alignment.

Yu Lin1, Shijie Jia1

  • 1School of Electrical Engineering, Dalian Jiaotong University, Dalian 116028, China.

Sensors (Basel, Switzerland)
|June 12, 2026
PubMed
Summary

SpecBEV improves multi-camera 3D object detection for autonomous driving by addressing illumination and occlusion issues. This novel framework enhances detection accuracy and localization stability using frequency-prior attention and cross-view alignment.

Keywords:
BEV object detectioncross-view feature alignmentfrequency priormulti-view fusionspatial attention

More Related Videos

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments
03:31

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments

Published on: December 15, 2023

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish
10:56

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish

Published on: March 6, 2014

Related Experiment Videos

Last Updated: Jun 13, 2026

Long-term Video Tracking of Cohoused Aquatic Animals: A Case Study of the Daily Locomotor Activity of the Norway Lobster (Nephrops norvegicus)
05:57

Long-term Video Tracking of Cohoused Aquatic Animals: A Case Study of the Daily Locomotor Activity of the Norway Lobster (Nephrops norvegicus)

Published on: April 8, 2019

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments
03:31

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments

Published on: December 15, 2023

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish
10:56

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish

Published on: March 6, 2014

Area of Science:

  • Computer Vision
  • Autonomous Driving Systems
  • Machine Learning

Background:

  • Multi-camera perception is a cost-effective alternative to LiDAR for autonomous driving.
  • Existing bird's-eye-view (BEV) detectors face challenges with illumination variations, occlusions, and cross-view inconsistencies.
  • These issues lead to background noise, geometric misalignment, missed detections, and localization instability.

Purpose of the Study:

  • To propose SpecBEV, an enhanced multi-view 3D object detection framework.
  • To mitigate challenges in feature projection and fusion for BEV representations.
  • To improve detection accuracy and localization stability in autonomous driving systems.

Main Methods:

  • Introduced a frequency-prior spatial attention module (SA-Freq) using discrete cosine transform (DCT) bases.
  • Developed a cross-view feature alignment module (CFA) for feature consistency.
  • Implemented an end-to-end BEV detection framework leveraging these modules.

Main Results:

  • SpecBEV achieved 0.3856 mAP and 0.4871 NDS on the nuScenes dataset.
  • Demonstrated significant improvements over the BEVDet baseline, with a 36.35% relative gain in mAP and 39.17% in NDS.
  • Validated the effectiveness of SA-Freq and CFA in suppressing redundant activations and reducing geometric inconsistency.

Conclusions:

  • SpecBEV effectively addresses illumination variations and cross-view inconsistencies in multi-camera 3D object detection.
  • The proposed framework enhances detection performance and localization stability for autonomous driving.
  • Frequency-prior attention and cross-view alignment are crucial for robust BEV perception.