.png){kind=icon}
Return
Camera Pose Estimation and Relocalization Algorithm Based on Ground Texture
.png){kind=icon}
PRE
AI
DOI：10.1109/TIM.2024.3428648.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In recent years, mobile robots have become progressively more prevalent in daily life. Throughout their usage, the accuracy and robustness of the system can be greatly affected by changes in the peripheral environment. However, the system has not fully utilized the available ground information. This article presents a solution that utilizes ground texture information to achieve accurate mapping and localization. The proposed solution encompasses the development of a comprehensive system that incorporates plane map construction, pose estimation, and automatic relocalization functions. The system effectively optimizes pose and mapping effects by incorporating artificial loop frames. Furthermore, it successfully achieves the automatic relocalization with a mismatch rate of less than 1% through the use of feature matching based on a prior map. This effectively resolves the issue of losing track of the target. This article conducts experiments on various ground texture features to thoroughly validate the accuracy and robustness of the system.
Keywords:
Feature extraction
Accuracy
Cameras
Pose estimation
Robustness
Location awareness
Three-dimensional displays
Ground texture
map saving
pose
relocalization
Journal
IF:
5.9
Papers:
2.0W
Citations:
5.8W
Organization
Cited Papers
.png){kind=icon}
ShareSaveShareSaveShareSaveSupporting better practice benchmarking: A DEA-ANN approach to bank branch performance assessment
OAAI
OAAIShareSaveShareSaveShareSave
View more