.png){kind=icon}
Return
Fractal Knowledge-Guided Diffusion Representation Learning Framework for Self-supervised Crack Segmentation
.png){kind=icon}
PRE
AI
DOI：10.1016/j.knosys.2026.116085.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Fractal pseudo-masks with topology guides the learning of unlabeled data. • Multi-scale gated diffusion module improves crack feature modeling. • Dual-stream spatially adaptive denormalization stabilizes adversarial training. • Contrastive unpaired translation achieves contrastive representation learning.
Keywords:
Fractal pseudo-masks
Multi-scale gated diffusion
Dual-stream spatially adaptive denormalization
Contrastive unpaired translation
Self-supervised crack segmentation
Journal
K
IF:
7.6
Papers:
1.3W
Citations:
4.5W
Organization
Cited Papers
ShareSavePixel-Level Cracking Detection on 3D Asphalt Pavement Images Through Deep-Learning- Based CrackNet-V
PREAI
ShareSaveSelf-supervised anomaly detection in computer vision and beyond: A survey and outlook.png)
NEURAL NETWORKS
OAAI
NEURAL NETWORKSIF6.3
OAAIShareSaveShareSaveDeepCrack: A deep hierarchical feature learning architecture for crack segmentationNEUROCOMPUTING
NEUROCOMPUTINGIF6.5
PREAI
ShareSave
View more