.png){kind=icon}
.png){kind=icon}
Return
Vision foundation model with connectivity-based inferencing for zero-shot bridge element segmentation and geometric reconstruction
.png){kind=icon}
PRE
AI
M
V
K
DOI：10.1016/j.autcon.2026.106995.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• SAM with connectivity-based semantic and geometric inferencing for bridge segmentation and reconstruction • Bridge-specific point-to-image projection for outdoor point cloud feature enhancement • Class-agnostic, annotation-free segmentation with rule-based filtering and cross-view label integration • Connectivity-based inferencing for semantic consistency and segmentation refinement • Over 98% instance-level mIoU on public dataset with connectivity-based segmentation refinement
Keywords:
bridge segmentation
geometric reconstruction
connectivity-based inferencing
zero-shot learning
point cloud enhancement
Journal
IF:
11.5
Papers:
6.1K
Citations:
4.2W
Organization
Cited Papers