Related Experiment Video
Updated: Apr 1, 2026
08:25
Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
Published on: May 7, 2019
9.7K
Biologically Motivated Novel Localization Paradigm by High-Level Multiple Object Recognition in Panoramic Images
1Advanced Visual Intelligence Laboratory, Yeungnam University, 280 Daehak-ro, Gyeongsan, Gyeongbuk 712-749, Republic of Korea.
Thescientificworldjournal
|October 13, 2015
Summary
This study introduces a new global localization method inspired by human visual systems (HVSs). It uses object recognition from panoramic images for accurate robot positioning and navigation.
Area of Science:
- Robotics
- Computer Vision
- Artificial Intelligence
Background:
- Human visual systems (HVSs) leverage object recognition for spatial awareness and navigation.
- Existing global localization methods often require extensive environmental mapping or sensor data.
Purpose of the Study:
- To propose a novel global localization method inspired by human visual systems (HVSs).
- To utilize object recognition from panoramic images for metric global localization (position and viewing direction).
- To enable robot-object interactions through high-level object information.
Main Methods:
- Panoramic image acquisition.
- Multiple object recognition from acquired images.
- Grid-based localization using bearing information of recognized objects.
Main Results:
- Demonstrated effective global localization using object recognition data.
- Achieved metric localization (position and viewing direction) from a single panoramic image.
- Validated the feasibility of the HVS-inspired localization paradigm experimentally.
Conclusions:
- The proposed HVS-inspired method offers a viable approach for global localization.
- Object recognition provides robust cues for both localization and robot-object interaction.
- This paradigm enhances robotic navigation capabilities by mimicking human visual strategies.
