DeltaSeg: Tiered Attention and Deep Delta Learning for Multi-Class Structural Defect Segmentation

• URL: http://arxiv.org/abs/2604.18745v1
• Date: Mon, 20 Apr 2026 18:49:16 GMT
• Title: DeltaSeg: Tiered Attention and Deep Delta Learning for Multi-Class Structural Defect Segmentation
• Abstract summary: This paper presents a U-shaped encoder-decoder architecture with a tiered attention strategy that integrates Squeeze-and-Excitation (SE) channel attention in the encoder, Coordinate Attention at the bottleneck and decoder, and a novel Deep Delta Attention (DDA) mechanism in the skip connections.<n>Across both benchmarks, DeltaSeg consistently outperforms 12 competing architectures including U-Net, SA-UNet, UNet3+, SegFormer, Swin-UNet, EGE-UNet, FPN, and Mobile-UNETR.
• Score: 1.2904450317782683
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Automated segmentation of structural defects from visual inspection imagery remains challenging due to the diversity of damage types, extreme class imbalance, and the need for precise boundary delineation. This paper presents DeltaSeg, a U-shaped encoder-decoder architecture with a tiered attention strategy that integrates Squeeze-and-Excitation (SE) channel attention in the encoder, Coordinate Attention at the bottleneck and decoder, and a novel Deep Delta Attention (DDA) mechanism in the skip connections. The encoder uses depthwise separable convolutions with dilated stages to maintain spatial resolution while expanding the receptive field. Atrous Spatial Pyramid Pooling (ASPP) at the bottleneck captures multi-scale context. The DDA module refines skip connections through a dual-path scheme combining a learned delta operator for nuisance feature suppression with spatial attention gates conditioned on decoder signals. Deep supervision through multi-scale auxiliary heads further strengthens gradient flow and encourages semantically meaningful features at intermediate decoder stages. We evaluate DeltaSeg on two datasets: the S2DS dataset (7 classes) and the Culvert-Sewer Defect Dataset (CSDD, 9 classes). Across both benchmarks, DeltaSeg consistently outperforms 12 competing architectures including U-Net, SA-UNet, UNet3+, SegFormer, Swin-UNet, EGE-UNet, FPN, and Mobile-UNETR, demonstrating strong generalization across damage types, imaging conditions, and structural geometries.

Related papers

• Open-World Semantic Segmentation with Sensitivity Modeling [6.086784305572313]
  Open-world vision systems must recognize known categories and detect unseen or anomalous content.<n>We extend a dual-decoder baseline with a third, complementary decoder within a unified encoder-decoder design.<n>Our method improves anomaly segmentation and novel-class discovery while maintaining competitive closed-set accuracy.
  arXiv  Detail & Related papers  (2026-08-08T19:37:02Z)
• DCAU-Net: Differential Cross Attention and Channel-Spatial Feature Fusion for Medical Image Segmentation [2.2015188658021003]
  We propose DCAU-Net, a novel yet efficient segmentation framework.<n>First, a new Differential Cross Attention (DCA) is designed to compute the difference between two independent softmax attention maps.<n>Second, a Channel-Spatial Feature Fusion (CSFF) strategy is introduced to adaptively recalibrate features.
  arXiv  Detail & Related papers  (2026-03-10T11:37:10Z)
• SPG-CDENet: Spatial Prior-Guided Cross Dual Encoder Network for Multi-Organ Segmentation [5.970991208589063]
  We propose a novel two-stage segmentation paradigm designed to improve multi-organ segmentation accuracy.<n>Our SPG-CDENet consists of two key components: a spatial prior network and a cross dual encoder network.<n>The global encoder captures global semantic features from the entire image, while the local encoder focuses on features from the prior network.
  arXiv  Detail & Related papers  (2025-10-30T11:33:29Z)
• O2Former:Direction-Aware and Multi-Scale Query Enhancement for SAR Ship Instance Segmentation [0.3611754783778107]
  Instance segmentation of ships in synthetic aperture radar (SAR) imagery is critical for applications such as maritime monitoring, environmental analysis, and national security.<n> SAR ship images present challenges including scale variation, object density, and fuzzy target boundary.<n>We propose O2Former, a tailored instance segmentation framework that extends Mask2Former by fully leveraging the structural characteristics of SAR imagery.
  arXiv  Detail & Related papers  (2025-06-13T16:06:51Z)
• MSA-UNet3+: Multi-Scale Attention UNet3+ with New Supervised Prototypical Contrastive Loss for Coronary DSA Image Segmentation [8.850534640462081]
  We propose a Supervised Prototypical Contrastive Loss that fuses supervised and prototypical contrastive learning to enhance coronary DSA image segmentation.<n>We implement the proposed SPCL loss within an MSA-UNet3+: a Multi-Scale Attention-Enhanced UNet3+ architecture.<n> Experiments on a private coronary DSA dataset show that MSA-UNet3+ outperforms state-of-the-art methods.
  arXiv  Detail & Related papers  (2025-04-07T15:35:30Z)
• Triple-View Knowledge Distillation for Semi-Supervised Semantic Segmentation [54.23510028456082]
  We propose a Triple-view Knowledge Distillation framework, termed TriKD, for semi-supervised semantic segmentation. The framework includes the triple-view encoder and the dual-frequency decoder.
  arXiv  Detail & Related papers  (2023-09-22T01:02:21Z)
• Uncertainty Driven Bottleneck Attention U-net for Organ at Risk Segmentation [20.865775626533434]
  Organ at risk (OAR) segmentation in computed tomography (CT) imagery is a difficult task for automated segmentation methods. We propose a multiple decoder U-net architecture and use the segmentation disagreement between the decoders as attention to the bottleneck of the network. For accurate segmentation, we also proposed a CT intensity integrated regularization loss.
  arXiv  Detail & Related papers  (2023-03-19T23:45:32Z)
• BCS-Net: Boundary, Context and Semantic for Automatic COVID-19 Lung Infection Segmentation from CT Images [83.82141604007899]
  BCS-Net is a novel network for automatic COVID-19 lung infection segmentation from CT images. BCS-Net follows an encoder-decoder architecture, and more designs focus on the decoder stage. In each BCSR block, the attention-guided global context (AGGC) module is designed to learn the most valuable encoder features for decoder.
  arXiv  Detail & Related papers  (2022-07-17T08:54:07Z)
• The KFIoU Loss for Rotated Object Detection [115.334070064346]
  In this paper, we argue that one effective alternative is to devise an approximate loss who can achieve trend-level alignment with SkewIoU loss. Specifically, we model the objects as Gaussian distribution and adopt Kalman filter to inherently mimic the mechanism of SkewIoU. The resulting new loss called KFIoU is easier to implement and works better compared with exact SkewIoU.
  arXiv  Detail & Related papers  (2022-01-29T10:54:57Z)
• Crosslink-Net: Double-branch Encoder Segmentation Network via Fusing Vertical and Horizontal Convolutions [58.71117402626524]
  We present a novel double-branch encoder architecture for medical image segmentation. Our architecture is inspired by two observations: 1) Since the discrimination of features learned via square convolutional kernels needs to be further improved, we propose to utilize non-square vertical and horizontal convolutional kernels. The experiments validate the effectiveness of our model on four datasets.
  arXiv  Detail & Related papers  (2021-07-24T02:58:32Z)
• Do Generative Models Know Disentanglement? Contrastive Learning is All You Need [59.033559925639075]
  We propose an unsupervised and model-agnostic method: Disentanglement via Contrast (DisCo) in the Variation Space. DisCo achieves the state-of-the-art disentanglement given pretrained non-disentangled generative models, including GAN, VAE, and Flow.
  arXiv  Detail & Related papers  (2021-02-21T08:01:20Z)
• Suppress and Balance: A Simple Gated Network for Salient Object Detection [89.88222217065858]
  We propose a simple gated network (GateNet) to solve both issues at once. With the help of multilevel gate units, the valuable context information from the encoder can be optimally transmitted to the decoder. In addition, we adopt the atrous spatial pyramid pooling based on the proposed "Fold" operation (Fold-ASPP) to accurately localize salient objects of various scales.
  arXiv  Detail & Related papers  (2020-07-16T02:00:53Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.