S3DNet: Support-Driven Spatial–Semantic Decoupling for Few-Shot Strip-Steel Surface Defect Segmentation
- 1Gezhouba Hydropower Station, China Yangtze Power Co., Ltd., Yichang 443002, China
- 2Zhejiang Dahua Technology Co., Ltd., Hangzhou 310000, China
- 3School of Automation, Hangzhou Dianzi University, Hangzhou 310000, China
Abstract
Few-shot strip-steel surface defect segmentation requires a model to identify unseen defect categories and delineate their precise boundaries from only a few annotated support images. This remains challenging because strip-steel defects are often small, elongated, low-contrast, and easily confused with repetitive background textures. To address these challenges, we propose the Support-Driven Spatial–Semantic Decoupling Network (S3DNet), which preserves a frozen DINOv3 encoder while explicitly separating support-conditioned semantic adaptation from fine-grained spatial refinement. Specifically, a low-rank prototype adapter transforms the masked support prototype into an episode-conditioned semantic update of the query tokens, enabling the frozen representation to adapt to the defect category specified by the current support set. In parallel, a lightweight spatial adapter extracts multiscale local cues for recovering weak boundaries and thin structures. A spatial–semantic interaction module subsequently uses support–query semantic similarity to selectively gate these spatial cues before progressive decoding, thereby suppressing irrelevant texture responses while retaining defect-relevant details. Experiments on FSSD-12 demonstrate that S3DNet consistently outperforms representative CNN-based few-shot segmentation methods under both 1-shot and 5-shot settings, while the lightweight DINOv3-Small configuration achieves a favorable accuracy–efficiency trade-off. Ablation and qualitative analyses further validate the effectiveness and complementarity of the proposed semantic adaptation, spatial refinement, and interaction mechanisms.
Related
- Active Learning under Simulated Class Imbalance and Label Shift in Binary Semantic Segmentation
- RePAIR: Inverting Backdoor Attacks in Semantic Segmentation
- Anatomical Network Analyses of Pre‐ and Post‐Metamorphic Cranial Systems Support the Adaptive Decoupling Hypothesis in Anura
- Covert Digital Semantic Communication With Movable Antennas: Constellation Shaping, Spatial Optimization, and Semantic Mapping
- Spatial Token Interleaving and Two-Pass BP Decoding for Semantic Image Transmission
BibTeX
@article{zhao2026s3dnet,
title = {{S3DNet: Support-Driven Spatial–Semantic Decoupling for Few-Shot Strip-Steel Surface Defect Segmentation}},
author = {Ju Zhao and Jingguang Yang and Fengling Lei and Baitong Wang and Zhiguo Li and Jiacheng Huang and Xiaofei Zhou},
journal = {Mathematics},
year = {2026},
month = oct,
doi = {10.3390/math14203730},
url = {https://www.mdpi.com/2227-7390/14/20/3730},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - S3DNet: Support-Driven Spatial–Semantic Decoupling for Few-Shot Strip-Steel Surface Defect Segmentation
AU - Ju Zhao
AU - Jingguang Yang
AU - Fengling Lei
AU - Baitong Wang
AU - Zhiguo Li
AU - Jiacheng Huang
AU - Xiaofei Zhou
PY - 2026
DA - 2026/10/14
JO - Mathematics
DO - 10.3390/math14203730
UR - https://www.mdpi.com/2227-7390/14/20/3730
AB - Few-shot strip-steel surface defect segmentation requires a model to identify unseen defect categories and delineate their precise boundaries from only a few annotated support images. This remains challenging because strip-steel defects are often small, elongated, low-contrast, and easily confused with repetitive background textures. To address these challenges, we propose the Support-Driven Spatial–Semantic Decoupling Network (S3DNet), which preserves a frozen DINOv3 encoder while explicitly separating support-conditioned semantic adaptation from fine-grained spatial refinement. Specifically, a low-rank prototype adapter transforms the masked support prototype into an episode-conditioned semantic update of the query tokens, enabling the frozen representation to adapt to the defect category specified by the current support set. In parallel, a lightweight spatial adapter extracts multiscale local cues for recovering weak boundaries and thin structures. A spatial–semantic interaction module subsequently uses support–query semantic similarity to selectively gate these spatial cues before progressive decoding, thereby suppressing irrelevant texture responses while retaining defect-relevant details. Experiments on FSSD-12 demonstrate that S3DNet consistently outperforms representative CNN-based few-shot segmentation methods under both 1-shot and 5-shot settings, while the lightweight DINOv3-Small configuration achieves a favorable accuracy–efficiency trade-off. Ablation and qualitative analyses further validate the effectiveness and complementarity of the proposed semantic adaptation, spatial refinement, and interaction mechanisms.
SN - 2227-7390
ER -