Synapse
⌘+K
Synapse
PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
February 27, 2026Open Access

Efficient person detection with LiDAR sensors for mobile robots

View Full Paper
Ask AI
Bookmark
Share

Authors

DJDan Jia

Discussion

Loading...

Member takes

Overview

Thesis demonstrates enhanced person detection in mobile robots using LiDAR, suggesting efficient sensor implementation.

Key Points

  • The aim is to design efficient person detection algorithms utilizing LiDAR sensors for mobile robots.
  • Developed DR-SPAAM, a person detection algorithm for 2D LiDAR sensors.
  • Created pseudo-labels from calibrated camera and image-based detector to improve training.
  • Implemented Global Hierarchical Attention for efficient processing of point clouds.
  • DR-SPAAM achieved 70.3% AP on the DROW dataset with 87.2 FPS on a dedicated GPU.
  • Pseudo-label training significantly enhanced detection performance and generalization.
  • Comparison showed 2D LiDAR detection accuracy on par with 3D LiDAR for visible persons.

Cite This Study

Dan Jia (2025) studied this question.

synapsesocial.com/papers/69a1350eed1d949a99abe9dfhttps://doi.org/10.18154/rwth-2026-01134
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Point cloud-based multi-target 3D object detection using LiDAR sensor and deep learning2026 · 3 citations
  2. 2Object Detection in Point Clouds for Mobile Robots2026
  3. 3Advancing Point Cloud Perception: A Focus on People Detection2025
  4. 4HFSA-Net: A 3D Object Detection Network with Structural Encoding and Attention Enhancement for LiDAR Point Clouds2026
  5. 5Approaching Outside: Scaling Unsupervised 3D Object Detection from 2D Scene2024