Xiaoqiang Shi

Xiaoqiang Shi

Auto-generated profile, based on publicly available information

Xiaoqiang Shi

Shenyang Institute of Computing Technology, Chinese Academy of Sciences, Shenyang, China

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
17 October 2023 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
XSXiaoqiang Shi
Xiaoqiang Shi
ZYZhenyu Yin
Zhenyu Yin
Guangjie HanGuangjie Han
Guangjie Han
WLWenzhuo Liu
Wenzhuo Liu
LQLi Qin
Li Qin
SLShurui Li
Shurui Li
Although semantic segmentation methods have made remarkable progress so far, their long inference process limits their use in practical applications. Recently, some two-branch and three-branch real-time segmentation networks have been proposed to improve segmentation accuracy by adding branches to extract spatial or border information. For the design of extracting spatial information branches, preserving high-resolution features or adding segmentation loss to guide spatial branches are commonly used methods to extract spatial information. However, these approaches are not the most efficient. To solve the problem, we design the spatial information extraction branch as an AutoEncoder structure, which allows us to extract the spatial structure and features of the image during the encoding and decoding process of the AutoEncoder. Border, semantic and spatial information are all helpful for segmentation tasks, and efficiently fusing these three kinds of information can obtain better feature representation compared to the fusion of two types of information in the dual-branch network. However, existing three-branch networks have yet to explore this aspect deeply. Therefore, this paper designs a new three-branch network based on this starting point. In addition, we also propose a feature fusion module called the Unified Multi-Feature Fusion module (UMF), which can fuse multiple features efficiently. Our method achieves a state-of-the-art trade-off between inference speed and accuracy on the Cityscapes, CamVid, and NightCity datasets. Specifically, BSSNet-T achieves 78.8% mIoU at 115.8 FPS on the Cityscapes dataset, 79.5% mIoU at 170.8 FPS on the CamVid dataset, and 52.6% mIoU at 172.3 FPS on the NightCity dataset. Code is available at https://github.com/SXQ-STUDY/BSSNet.

15 July 2024 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
ZYZhenyu Yin
Zhenyu Yin
LQLi Qin
Li Qin
Guangjie HanGuangjie Han
Guangjie Han
XSXiaoqiang Shi
Xiaoqiang Shi
FZFeiqing Zhang
Feiqing Zhang
YBYuanguo Bi
Yuanguo Bi
Semantic segmentation of surface defects is essential to ensure product quality in intelligent manufacturing. However, due to the diversity and complexity of industrial scenarios and defects, existing defect semantic segmentation methods still suffer from inconsistent intraclass and indistinguishable interclass segmentation results. To overcome these problems, we propose a new dual-branch surface defect semantic segmentation network, DDSNet. First, we integrate semantic and border information to enrich the feature representation of defects and solve the problem of indistinguishable interclass segmentation results. Next, we introduce a global and local feature fusion (GLF) module based on similarity metrics to guide the network in further refining and highlighting the detail feature on defects to solve the problem of inconsistent intraclass segmentation results. In addition, to enrich the surface defect segmentation datasets, we collect datasets of steel foil surface defects, Ste-Seg, and aluminum block surface defects, Alu-Seg. Experimental results for five datasets of semantic segmentation of defects show that DDSNet outperforms the state-of-the-art methods in terms of mIoU (NEU-Seg: 85.12%, MT-Defect: 76.51%, MSD: 91.82%, Ste-Seg: 90.01%, and Alu-Seg: 84.77%). All our experiments were conducted on a NVIDIA GTX 3060Ti. The dataset and code are available at https://github.com/QinLi-STUDY/DDSNet .

27 September 2025 in Engineering Applications of Artificial Intelligence
Verified
Journal Article
Closed access
LQLi Qin
Li Qin
ZYZhenyu Yin
Zhenyu Yin
FZFeiqing Zhang
Feiqing Zhang
CSChunhe Song
Chunhe Song
XSXiaoqiang Shi
Xiaoqiang Shi

26 February 2026 in Engineering Applications of Artificial Intelligence
Verified
Journal Article
Closed access
ZYZhenyu Yin
Zhenyu Yin
LQLi Qin
Li Qin
FZFeiqing Zhang
Feiqing Zhang
YYYouwei Yu
Youwei Yu
XSXiaoqiang Shi
Xiaoqiang Shi
YBYuanguo Bi
Yuanguo Bi

3 January 2026 in Pattern Recognition
Verified
Journal Article
Closed access
ZYZhenyu Yin
Zhenyu Yin
XSXiaoqiang Shi
Xiaoqiang Shi
FZFeiqing Zhang
Feiqing Zhang
DLDong Li
Dong Li
LNLin Nie
Lin Nie
LQLi Qin
Li Qin

27 April 2026 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
ZYZhenyu Yin
Zhenyu Yin
LYLijuan Yuan
Lijuan Yuan
FZFeiqing Zhang
Feiqing Zhang
XSXiaoqiang Shi
Xiaoqiang Shi
LQLi Qin
Li Qin
High-precision surface defect inspection is critical for ensuring product quality in precision parts manufactured by computer numerical control (CNC) machining. It can also provide quantitative measurement outputs for quality assessment. Given the limited availability of anomalous samples and the diversity of part categories, unsupervised anomaly detection (UAD) applies to this inspection task. However, existing multi-class UAD methods are susceptible to feature distribution heterogeneity and cross-class semantic interference during joint modeling, which degrades both defect detection accuracy and the stability of quantitative measurement outputs. To address this issue, we propose CNC-Former, a Vision Transformer-based high-precision multi-class UAD framework. First, a semantic-enhanced linear attention is introduced to suppress irrelevant cross-class responses by constraining the attention distribution, thereby improving defect localization performance. Second, an adaptive dual-branch bottleneck is constructed to jointly model semantic and structural information, enabling more effective perception of subtle local defects and their boundaries. Third, a progressive feature fusion module is incorporated to enhance cross-layer feature consistency and reduce inter-class interference. In addition, a measurement-oriented industrial inspection instrument for precision mechanical parts has been developed. The system converts the pixel-level outputs of CNC-Former into a metrologically defined projected defect area, thereby establishing a traceable end-to-end measurement chain from image acquisition to quantitative defect assessment. Experimental results show that, across 12 categories in a CNC-machined precision-part surface defect dataset, CNC-Former achieves a mean image-level AUROC of 94.2% and a pixel-level AUROC of 90.8%. Meanwhile, the measurement uncertainty of the projected defect area is evaluated, and the system output’s stability is validated through repeatability and reproducibility experiments.