Shu Wang

Prof. Shu Wang

orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

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

Representative Publications

View All Publications
30 September 2025 in Electronics
Verified
Journal Article
Open access
PDF file
SWShu Wang
Shu Wang
ATAdriano Tavares
Adriano Tavares
CLCarlos Santos Lima
Carlos Santos Lima
TGTiago Gomes
Tiago Gomes
YZYicong Zhang
Yicong Zhang
YLYanchun Liang
Yanchun Liang
Posture recognition is critical in modern educational and office environments for preventing musculoskeletal disorders and maintaining cognitive performance. Existing methods based on human keypoint detection typically rely on convolutional neural networks (CNNs) and single-scale features, which limit representation capacity and suffer from overfitting under small-sample conditions. To address these issues, we propose MSBN-SPose, a Multi-Scale Bayesian Neuro-Symbolic Posture Recognition framework that integrates geometric features at multiple levels—including global body structure, local regions, facial landmarks, distances, and angles—extracted from OpenPose keypoints. These features are processed by a multi-branch Bayesian neural architecture that models epistemic uncertainty, enabling improved generalization and robustness. Furthermore, a lightweight neuro-symbolic reasoning module incorporates human-understandable rules into the inference process, enhancing transparency and interpretability. To support real-world evaluation, we construct the USSP dataset, a diverse, classroom-representative collection of student postures under varying conditions. Experimental results show that MSBN-SPose achieves 96.01% accuracy on USSP, outperforming baseline and traditional methods under data-limited scenarios.

6 February 2026 in IEEE Access
Verified
Journal Article
Open access
YZYicong Zhang
Yicong Zhang
SWShu Wang
Shu Wang
RMRui Ma
Rui Ma
XDXinzheng Dong
Xinzheng Dong
YLYu Li
Yu Li
Target detection in Synthetic Aperture Radar (SAR) imagery is consistently hampered by speckle noise, geometric distortions, and complex scattering characteristics. Existing detection methods often remain confined to the spatial domain, struggling to capture global structural information and effectively leverage the rich data present in the frequency domain. To address these limitations, this paper introduces FMD-Net (Frequency, Multi-dimensional, Dynamic Net), a detection framework centered on the deep integration of frequency domain analysis with an advanced attention-based architecture. Frequency domain analysis allows for the extraction of periodic textures and global structures while simultaneously suppressing noise, compensating for the limitations of purely spatial-domain approaches. FMD-Net introduces a set of synergistic modules that dynamically filter and enhance features in the frequency space, adaptively modulate multi-dimensional features to handle scale and shape variations, and meticulously recover fine details for small targets through content-aware upsampling. FMD-Net achieves a mean Average Precision (mAP50-95) of 68.9% on the SHSA dataset. Furthermore, it attains a competitive (mAP50-95) of 67.6% on the challenging MSAR dataset and 63.6% on the FUSAR-Ship dataset. These results suggest that FMD-Net demonstrates competitive performance against state-of-the-art methods, showing noticeable improvements in detection accuracy and robustness, especially for small targets and within complex, cluttered scenes.

25 November 2025 in Entropy
Verified
Journal Article
Open access
PDF file
SWShu Wang
Shu Wang
ATAdriano Tavares
Adriano Tavares
CLCarlos Santos Lima
Carlos Santos Lima
TGTiago Gomes
Tiago Gomes
YZYicong Zhang
Yicong Zhang
YLYanchun Liang
Yanchun Liang
Sitting posture recognition, defined as automatically localizing and categorizing seated human postures, has become essential for large-scale ergonomics assessment and longitudinal health-risk monitoring in classrooms and offices. However, in real-world multi-person scenes, pervasive occlusions and overlaps induce keypoint misalignment, causing global-attention backbones to fail to localize critical local structures. Moreover, annotation scarcity makes small-sample training commonplace, leaving models insufficiently robust to misalignment perturbations and thereby limiting cross-domain generalization. To address these challenges, we propose LAViTSPose, a lightweight cascaded framework for sitting posture recognition. Concretely, a YOLOR-based detector trained with a Range-aware IoU (RaIoU) loss yields tight person crops under partial visibility; ESBody suppresses cross-person leakage and estimates occlusion/head-orientation cues; a compact ViT head (MLiT) with Spatial Displacement Contact (SDC) and a learnable temperature (LT) mechanism performs skeleton-only classification with a local structural-consistency regularizer. From an information-theoretic perspective, our design enhances discriminative feature compactness and reduces structural entropy under occlusion and annotation scarcity. We conducted a systematic evaluation on the USSP dataset, and the results show that LAViTSPose outperforms existing methods on both sitting posture classification and face-orientation recognition while meeting real-time inference requirements.

Verified
Journal Article
Open access
PDF file
CZChulei Zhang
Chulei Zhang
RLRuikai Liu
Ruikai Liu
YLYuanshu Li
Yuanshu Li
XWXuan Wu
Xuan Wu
GLGang Lin
Gang Lin
YZYou Zhou
You Zhou
Pseudo-label-based deep clustering has emerged as a compelling paradigm for unsupervised visual learning, yet two fundamental limitations impede its effectiveness: (1) pseudo-label noise, where unreliable hard assignments generated in early training accumulate and propagate errors across iterations; and (2) confidence-insensitive pseudo-label supervision, where one-hot pseudo-labels are treated with equal importance despite substantial differences in their reliability, increasing the influence of uncertain samples, particularly those near cluster boundaries. To address both issues simultaneously, we propose PSPLC—a Progressive Sample-Weighted Pseudo-Label Clustering framework with three synergistic modules. First, a MoCo-based contrastive pre-training module extracts discriminative feature representations. Second, a progressive sample-weighted pseudo-label training module (the core contribution) iterates an expectation step—which progressively expands a high-confidence sample set outward from calibrated cluster centres and assigns one-hot pseudo-labels with continuous confidence weights derived from a global–local confidence mechanism—with a maximisation step that jointly optimises a clustering consistency loss, a confidence-weighted one-hot pseudo-label loss, and an entropy balance loss. Third, a semi-supervised boosting module (PSPLC-Boost) identifies reliable samples via nearest-neighbour semantic consistency and jointly fine-tunes the feature extractor and clustering head with a FixMatch-inspired strategy. Extensive experiments on five benchmarks (STL-10, CIFAR-10, CIFAR-20, ImageNet-10, ImageNet-Dogs-15) show that PSPLC achieves state-of-the-art performance among purely unsupervised methods, and PSPLC-Boost surpasses all competitors on all five datasets across all three metrics.