mlsanityData to 14 Oct 2026

FORTRESS: Function-composition Optimized Real-Time Resilient Structural Segmentation via Kolmogorov-Arnold Enhanced Spatial Attention Networks

arXiv 2507.12675 ↗First posted 16 Jul 2025cs.CV cs.AI eess.IVcited by 0

Not on the map. Why?

Abstract (arXiv)

Automated structural defect segmentation in civil infrastructure faces a critical challenge: achieving high accuracy while maintaining computational efficiency for real-time deployment. This paper presents FORTRESS (Function-composition Optimized Real-Time Resilient Structural Segmentation), a new architecture that balances accuracy and speed by using a special method that combines depthwise separable convolutions with adaptive Kolmogorov-Arnold Network integration. FORTRESS incorporates three key innovations: a systematic depthwise separable convolution framework achieving a 3.6x parameter reduction per layer, adaptive TiKAN integration that selectively applies function composition transformations only when computationally beneficial, and multi-scale attention fusion combining spatial, channel, and KAN-enhanced features across decoder levels. The architecture achieves remarkable efficiency gains with 91% parameter reduction (31M to 2.9M), 91% computational complexity reduction (13.7 to 1.17 GFLOPs), and 3x inference speed improvement while delivering superior segmentation performance. Evaluation on benchmark infrastructure datasets demonstrates state-of-the-art results with an F1- score of 0.771 and a mean IoU of 0.677, significantly outperforming existing methods including U-Net, SA-UNet, and U- KAN. The dual optimization strategy proves essential for optimal performance, establishing FORTRESS as a robust solution for practical structural defect segmentation in resource-constrained environments where both accuracy and computational efficiency are paramount. Comprehensive architectural specifications are provided in the Supplemental Material. Source code is available at URL: https://github.com/faeyelab/fortress-paper-code.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 25 of its 59 references that are in the map. First posted on arXiv 16 Jul 2025.

Builds on

25 of 59 references.

References · 25

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  2. 2016Feature Pyramid Networks for Object Detectioncited by 5,036
  3. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,859
  4. 2021SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformerscited by 2,940
  5. 2018UNet++: A Nested U-Net Architecture for Medical Image Segmentationcited by 1,525
  6. 2018Attention U-Net: Learning Where to Look for the Pancreascited by 1,435
  7. 2021Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentationcited by 1,222
  8. 2019EfficientDet: Scalable and Efficient Object Detectioncited by 1,020
  9. 2024KAN: Kolmogorov-Arnold Networkscited by 649
  10. 2020UNet 3+: A Full-Scale Connected UNet for Medical Image Segmentationcited by 332
  11. 2022UNeXt: MLP-based Rapid Medical Image Segmentation Networkcited by 185
  12. 2024U-KAN Makes Strong Backbone for Medical Image Segmentation and Generationcited by 179
13 more
  1. 2020SA-UNet: Spatial Attention U-Net for Retinal Vessel Segmentationcited by 71
  2. 2024A Survey on Kolmogorov-Arnold Networkcited by 68
  3. 2023FasterViT: Fast Vision Transformers with Hierarchical Attentioncited by 65
  4. 2023EGE-UNet: an Efficient Group Enhanced UNet for skin lesion segmentationcited by 53
  5. 2024KANICE: Kolmogorov-Arnold Networks with Interactive Convolutional Elementscited by 12
  6. 2024Kolmogorov-Arnold Network Autoencoderscited by 8
  7. 2023Dual Attention U-Net with Feature Infusion: Pushing the Boundaries of Multiclass Defect Segmentationcited by 6
  8. 2024MobileUNETR: A Lightweight End-To-End Hybrid Vision Transformer For Efficient Medical Image Segmentationcited by 6
  9. 2024A Glass-Box Deep-Learning Method for Electrical Energy System Modeling Based on Kolmogorov-Arnold Networkcited by 2
  10. 2023GAEI-UNet: Global Attention and Elastic Interaction U-Net for Vessel Image Segmentationcited by 1
  11. 2024Dynamic Label Injection for Imbalanced Industrial Defect Segmentationcited by 1
  12. 2024KAT to KANs: A Review of Kolmogorov-Arnold Networks and the Neural Leap Forwardcited by 1
  13. 2025Kolmogorov-Arnold networks for metal surface defect classificationcited by 1

Built on by

Not cited yet.