In continuous sign language recognition (CSLR), existing methods predominantly adopt frame-wise feature extraction, neglecting temporal continuity and motion trajectory modeling, thereby struggling to capture complete spatio-temporal dynamics. Meanwhile, global cross-modal attention approaches typically directly model interactions between text and the entire video sequence but lack structural constraints, making them susceptible to interference from redundant frames, which leads to attention distribution dilution and undermines fine-grained motion alignment capability. To address these issues, this paper proposes CMST-Net, a Cross-modal Interaction and Spatio-temporal Feature Enhancement Method for Continuous Sign Language Recognition. CMST-Net comprises two principal modules: the Local–Global Cross-modal Fusion Module (LGCFM) and the Spatio-Temporal Feature Enhancement Module (STFEM). LGCFM introduces a synergistic modeling mechanism that combines local sliding-window attention with global attention, achieving a unified fusion of structured local alignment and global semantic modeling. STFEM incorporates multi-scale spatial dilated convolution and coordinate attention to extract fine-grained spatial features while leveraging channel partitioning and a hierarchical residual structure to enhance long-range temporal modeling capability; the two modules collaboratively yield high-quality spatio-temporal feature representations. Experiments on three public benchmark datasets (PHOENIX2014, PHOENIX2014-T, and CSL-Daily) demonstrate that CMST-Net can effectively improve continuous sign language recognition performance, achieving state-of-the-art performance on the PHOENIX2014 and CSL-Daily datasets and competitive results on the PHOENIX2014-T dataset.

In complex environments, sign language recognition (SLR) is easily affected by background clutter, motion blur, and rapid movement. These factors can obscure subtle gesture patterns and weaken the discriminative spatiotemporal features required for recognition. To address these challenges, we propose SST-Net, a spatiotemporal collaborative attention enhancement framework for SLR. First, CGSA introduces input-dependent gates to dynamically fuse spatial and channel attention, rather than using fixed fusion weights. Second, the TFM and SFM both exploit inter-frame differences but use them differently: the TFM enhances motion-sensitive temporal features, whereas the SFM suppresses motion-dominated components to preserve spatial details; both are further refined through feature reweighting. Third, experiments on CSL-100, INCLUDE, and AUTSL show that SST-Net improves accuracy by 2–4 percentage points over strong baselines while using 35 M parameters and 115 GFLOPs, demonstrating a competitive balance between recognition performance and computational cost.

Video action recognition requires the joint modeling of spatial appearance information and temporal dynamics. However, existing efficient action recognition methods based on two-dimensional convolution still have limitations in representing multi-scale spatial cues and aggregating key spatiotemporal information. To address these issues, this paper proposes an action recognition network based on Multi-Scale Dilated Feature Fusion and Decoupled Spatiotemporal Attention Pooling, termed MDSTA-Net. The proposed method adopts TSM as the basic temporal modeling framework and ResNet-50 as the backbone network. First, a Multi-Scale Dilated Feature Fusion module (MSDF) is designed to construct continuous multi-scale receptive fields through parallel convolutional branches with different dilation rates. An adaptive branch aggregation mechanism is further introduced to dynamically fuse responses at different scales, thereby enhancing the representation of both local details and broader contextual information. Second, a Decoupled Spatiotemporal Attention Pooling module (DSTAP) is proposed to model key action frames along the temporal dimension and salient discriminative regions along the spatial dimension. A residual pooling path is also incorporated to preserve global semantic information, improving the discriminative capability of video-level action representations. Experimental results on three public datasets, namely Something-Something V2, Kinetics-400, and HMDB51, demonstrate that MDSTA-Net achieves favorable recognition performance compared with several representative methods. Ablation studies further verify the effectiveness of MSDF and DSTAP, indicating that multi-scale spatial feature enhancement and key spatiotemporal information aggregation can effectively improve action recognition performance.

Human pose estimation in factory surveillance is challenged by scale variation, limb occlusion, complex backgrounds, and spatial detail loss during upsampling. This paper proposes MFA-Pose, an improved YOLO11s-Pose framework that enhances contextual representation and cross-scale feature reconstruction. The Multi-Scale Context Fusion (MSCF) module preserves directional positional information and models multi-range cross-channel dependencies using parallel one-dimensional convolutions with kernel sizes of 3, 5, and 7. The complete C2MSCF module contains approximately 0.790 M learnable parameters, compared with approximately 0.991 M parameters in the original C2PSA module. The Multi-Receptive-Field Adaptive Gated Upsampling (MRAG) module predicts sampling offsets through standard and dilated convolution branches, suppresses unreliable offsets using gating mechanisms, and combines dynamic resampling with a stable bilinear interpolation reference. The two MRAG modules used in MFA-Pose contain approximately 0.457 M and 0.130 M learnable parameters, respectively, corresponding to approximately 0.587 M parameters in total. On COCO 2017, MFA-Pose achieves 89.0% ${\mathrm{AP}}_{50}^{\mathrm{pose}}$ and 61.7% ${\mathrm{AP}}_{50:95}^{\mathrm{pose}}$, outperforming YOLO11s-Pose by 2.7 and 1.6 percentage points, respectively, with only 0.40 M additional parameters. Quantitative evaluation on the annotated industrial surveillance test set further shows that MFA-Pose achieves 87.2% Precision, 77.0% Recall, 81.8% F1-score, 88.7% ${\mathrm{AP}}_{50}^{\mathrm{pose}}$, and 61.1% ${\mathrm{AP}}_{50:95}^{\mathrm{pose}}$, outperforming the YOLO11s-Pose baseline across all evaluated accuracy metrics. Qualitative comparisons further show more coherent pose predictions under self-occlusion, non-standard working postures, and cluttered equipment backgrounds. Overall, MFA-Pose provides a favorable balance between pose estimation accuracy and model complexity for industrial surveillance applications.