Comprehending occluded objects remains a critical challenge for multi-modal large language models due to missing visual representations. Current approaches either rely on multi-stage pipelines with error propagation, or use unified encoders that fail to balance understanding and generation. We propose OCC-MLLM-V2, an end-to-end autoregressive framework with three key innovations: (1) Hierarchical Trinity Fusion Architecture integrating multi-view RGB, hand masks, and 3D reconstructions via Adaptive Weight Image Fusion and Spatial Attention Affine Fusion; (2) Visual Dual Encoder employing SigLIP for understanding and VQ tokenizers for generation; (3) Visual Dual Decoder with joint optimization. Unlike pipeline methods, our unified framework eliminates sequential dependencies and enables end-to-end gradient flow. Experiments demonstrate improvements: 6.27% gains on ObMan and 6.35% on DexYCB across 1B-8B models, while reducing FLOPs by 32%–59% and inference time by 9.5%–37%. Our framework surpasses GPT-4o and Gemini 2.5-Pro on multiple occlusion benchmarks. Code and data are available at https://github.com/chaoyiwang09/OCC-MLLM .

The sub-Nyquist sampling technique offers the possibility of wideband spectrum sensing using portable devices. Therefore, it becomes more urgent and important to improve the hardware and computational efficiency of existing sparse recovery algorithms. In this paper, we propose an improved non-iterative joint support set recovery algorithm based on the MUSIC criterion for Modulated Wideband Converter (MWC) sub-Nyquist sampling front-ends. Through subspace analysis, we find that the MUSIC-based algorithm has the potential to reduce hardware requirements. Moreover, simulations show that the proposed improved MUSIC algorithm has comparable performance to the iterative SOMP in coarse spectrum sensing applications.

Wideband spectrum sensing based on sub-Nyquist sampling is an attractive approach to advance dynamic spectrum sharing (DSS), which can improve frequency resource utilization while overcoming sampling bottlenecks. Under the Compressive Sensing (CS) framework, finding occupied subbands can be equivalent to computing the support set for the Multiple Measurement Vectors (MMV) problem. To guarantee the performance of joint support recovery in noisy environments, different kinds of prior information are required, one of which is sparsity, a time-varying parameter. To address the dependence of recovery performance on signal sparsity, this paper proposed a two-step scheme for blind wideband spectrum sensing using a Modulated Wideband Converter (MWC) sub-Nyquist sampling front-end. The scheme first adopts the model order selection (MOS) method to estimate sparsity from the compressed covariance matrix, and then uses the estimates to dynamically adjust joint support recovery. The complete theoretical derivation innovatively applies MOS to sub-Nyquist sampling and presents a design method for MOS penalty constant. Extensive simulation results show that the proposed scheme can not only achieve blind sensing under the spectrum occupancy up to 40%, reduce the overall computational complexity of iterative SOMP, but also significantly improve the false alarm performance, meeting the requirements of the IEEE 802.22 standard.

Abstract Recently, 3D Convolutional Neural Networks (3D CNNs) have attracted extensive attention in extracting spatial and temporal features in videos for their efficient feature extraction ability. However, it also brings enormous model parameters by training very deep 3D CNNs. Here, a novel network named spurious‐3D Residual Attention Networks (S3D RANs) is proposed for video‐based action recognition, which has the powerful capacity to learn collaborative spatiotemporal features. In particular, by leveraging the merits from 2D Convolutional Neural Networks (2D CNNs) and 3D CNNs, 2D CNNs are applied rather than 3D CNNs on frames of the single view of volumetric videos data to learn temporal motion features directly. Furthermore, view and channel‐wise attention mechanism submodules are employed in the residual unit to learn the importance of each view for action recognition and guide the network to pay more attention to the more useful information for action recognition. Experimental results on UCF‐101, HMDB‐51 datasets demonstrate that our S3D RANs have higher accuracy and lower model complexity than existing works.

Liquid crystal display (LCD) defects detection on module level is increasingly important for flat-panel displays (FPD) industry to increase the production capacity via machine vision technology. However, it is an overwhelmingly challenging issue due to various difficulties. This article discloses a practical automatic optical inspection (AOI) system consisting of hardware structure and software algorithm to detect module-level defects. The AOI system is the core component to build a distributed integrated inspection system with the help of the Internet of Things (IoT). Starting from the analysis of the challenges encountered in module-level defects inspection, a delicate photograph scheme is proposed to reveal different kinds of defects. In order to robustly work on the module-level defects detection with complex situations, a novel framework based on YOLOV3 detection unit is proposed in this article, including the preprocessing module, detection module, defects definition module, and interferences elimination module. To the best of our knowledge, this is the first work that designs a practical AOI system for module-level defects detection. In order to demonstrate the effectiveness of the proposed method, extensive experiments have been conducted on the manufacturing lines. The evaluation of the detection performance of the AOI system in comparison with a manual scheme indicates that the proposed system is practical for module-level defects detection. Currently, the proposed system has been deployed in a real-world LCD manufacturing line from a major player in the world.

It has been observed that the Internet of Medical Things (IoMT) is being deployed to construct varieties of intelligent platforms in medical and healthcare field, in order to comprehensively improve the quality of medical services. However, the cyber security of IoMT is facing enormous threat. Although many trust schemes are proposed to address the issue, the ignorance of aging weight in trust increases the risk of long-term attacks before being detected. In this paper, we design a dynamic aging weight scheme for trust model in IoMT. Essentially, when there are cooperative behaviors between two nodes, the aging weight can be set as large as possible to slow down the increase in the trust value of normal nodes. Once noncooperative behaviors appear, the smaller aging weight could mitigate the danger of compromised nodes. The simulation results indicate that our proposed scheme could better meet the principle of “Easy to lose” for trust.

Object reidentification with the goal of matching pedestrian or vehicle images captured from different camera viewpoints is of considerable significance to public security. Quadruple directional deep learning features (QD-DLFs) can comprehensively describe object images. However, the correlation among QD-DLFs is an unavoidable problem, since QD-DLFs are learned with quadruple independent directional deep networks (QIDDNs) driven with the same training data, and each network holds the same basic deep feature learning architecture (BDFLA). The correlation among QD-DLFs is harmful to the complementarity of QD-DLFs, restricting the object reidentification performance. For that, we propose joint quadruple decorrelation directional deep networks (JQD3Ns) to reduce the correlation among the learned QD-DLFs. In order to jointly train JQD3Ns, besides the softmax loss functions, a parameter correlation cost function is proposed to indirectly reduce the correlation among QD-DLFs by enlarging the dissimilarity among the parameters of JQD3Ns. Extensive experiments on three publicly available large-scale data sets demonstrate that the proposed JQD3Ns approach is superior to multiple state-of-the-art object reidentification methods.

Seismic signal is widely used in ground target classification due to its inherent characteristics. However, its propagation is highly dependent on local underlying geology. It means that nearly every one geographical environment requires a unique classifier. To resolve the problem, this paper presents a robust feature extraction method Log-Sigmoid Frequency Cepstral Coefficients (LSFCC) which evolves from Mel frequency cepstral coefficients (MFCC) for ground target classification by means of geophone. With the LSFCCs, the average classification accuracy of tracked and wheeled vehicle is more than 89% in three different geographical environments by only one classifier which is trained in one of the three environments.