Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 2, 2026Computational Intelligence

SAM‐IND: Enhancing SAM With Implicit Neural Decoder for Structural Crack Detection

View Full Paper
Ask AI
Bookmark
Share

Authors

LZLingjun ZhaoBWBin WangHMHua Ma

Discussion

Loading...

Member takes

Overview

Randomized trial explores crack detection in infrastructure, highlighting a novel framework's effectiveness.

Key Points

  • This research aims to improve structural crack detection by enhancing the segment anything model (SAM) with implicit neural representations.
  • Developed the SAM-IND framework combining SAM with implicit neural decoder strategies.
  • Utilized low-rank adaptation (LoRA) matrices for domain-specific feature learning.
  • Conducted extensive experiments on three benchmark datasets to validate model performance.
  • Achieved an F1 score of 89.63% on the DeepCrack dataset.
  • Reduced cross-material generalization error by 8.74% compared to state-of-the-art methods.
  • Required only 1.59M trainable parameters, constituting 1.75% of SAM's total parameters.

Cite This Study

Zhao et al. (2026) studied this question.

synapsesocial.com/papers/6a45ff6f9ed134303130fe7chttps://doi.org/10.1111/coin.70272
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. 1Crack-SAM: Crack Segmentation Using a Foundation Model2024
  2. 2Segment anything model-based crack segmentation using low-rank adaption fine-tuning2024 · 2 citations
  3. 3A Two-Stage Concrete Crack Segmentation Method Based on the Improved YOLOv11 and Segment Anything Model2026
  4. 4Directional multi-scale CNN-transformer hybrid network for robust crack segmentation2026
  5. 5Automated Detection and Segmentation of Cracks in Urban Underground Structures Based on YOLOv8-SAM22026