Coordinate classification is an efficient approach to 2-D human pose estimation (HPE), treating keypoint predictions as sub-pixel bins along horizontal and vertical axes, thereby avoiding the computationally intensive upsampling process required in traditional heatmap-based methods. In this article, we introduce the Position Encoding Coordinate Classification (PECC) system, which enhances coordinate classification by embedding position information directly into keypoint feature representations through a novel position encoding mechanism. We further design a tailored attention mechanism, Filtering Amplified Attention (FAA), optimized for coordinate classification. FAA provides finer relative positional information, improving the system’s ability to model relationships between keypoints and enhancing coordinate localization accuracy. Our method maintains the efficiency of coordinate classification by utilizing 1-D vectors, significantly reducing model parameters and computational cost. Additionally, the incorporation of positional encoding enhances the system’s ability to effectively model and exploit spatial information within a coordinate-classification-based pose estimation framework. Extensive experiments on mainstream datasets demonstrate that PECC achieves superior accuracy and robustness in 2-D HPE, advancing the state-of-the-art in this domain.

Through Internet of Things (IoT) communication technology, collaborative perception enhances a vehicle’s capacity to discern its surroundings while driving by integrating and synchronizing sensor data from multiple agents. With the advancement of cooperative perception techniques in single-modality methods, there has been a growing trend toward integrating multimodal data from heterogeneous sensors in recent years. However, due to the data heterogeneity inherent in diverse sensors, Bird’s Eye View (BEV) maps generated from different types of sensors may exhibit local discrepancies in the spatial representation of entity positions. Furthermore, individual agents may produce uncertain and flawed feature representations in real noisy environments. The influence of this indeterminacy exacerbates the issue of local inconsistency, leading to misalignment of the detected target during BEV alignment and fusion, thereby reducing detection accuracy. To address these problems, we propose a modal decision-making spatial alignment cooperative perception network (MDNet). First, the network generates BEV feature maps through dense depth image supervision for voxel feature extraction and model-guided selective feature fusion. Subsequently, we achieve enhanced accuracy in object detection by performing spatial alignment of BEV representations generated from two distinct sensors, both globally and locally within the spatial domain. Besides, we employ a cascaded centralized pyramid strategy during the message fusion stage, facilitating flexible sampling across horizontal and vertical spatial dimensions, promoting deep interaction among multiple agents. We conduct quantitative and qualitative experiments on the public OPV2V and DAIR-V2X-C benchmarks, and our proposed MDNet exhibits superior performance and stronger robustness in the 3-D object detection task, providing more precise target detection results.

Visual surface anomaly detection targets the location of anomalies, with numerous methods available to address the challenge. Reconstruction-based methods are popular for their adaptability and interpretability. However, reconstruction-based methods currently struggle with the challenge of achieving low image fidelity and a tendency to reconstruct anomalies. To overcome these challenges, we introduces the Feature Bank-guided Reconstruction method (FBR), incorporating three innovative modules: anomaly simulation, feature bank module, and a cross-fused Discrete Cosine Transform channel attention module. Guided by these modules, our method is capable of reconstructing images with enhanced robustness. The experimental results validate the effectiveness of the proposed approach, which not only achieves outstanding performance on the BeanTech AD dataset with an 96.4% image-AUROC and a 97.3% pixel-AUROC, but also demonstrates competitive performance on the MVTec AD dataset with a 99.5% image-AUROC and a 98.3% pixel-AUROC.

The state-of-the-art YOLO detection algorithms still suffer from the issue of redundant extraction of similar features during feature propagation, and the simplistic stacking approach of connecting different features limits the flexibility of feature fusion. We propose a new feature recombination mechanism involving refining feature extraction and flexible concatenation. It includes the HFConv (Hybrid Flexibility Convolution) module, the MFD (Multivariate Flexibility Downsampling) module, and the DFSPP (Deformable and Flexible Spatial Pyramid Pooling) module. Specifically, the HFConv module employs feature refinement and flexible connection strategies to optimize feature representation and reduce redundancy in a dynamic way, acquiring diverse feature information from local and surrounding regions. The MFD module leverages multiple downsampling methods to address the issue of feature redundancy that may arise from a single downsampling method, thereby enhancing feature diversity. The DFSPP module learns an offset corresponding to the pooling kernel size, allowing for the extraction of the most critical information in a dynamic manner. By incorporating these modules into the YOLO architecture, we develop a more robust network called FRFCNet, and the experimental results show a notable 4.1% and 2.8% improvement in AP values on the VOC2012 and COCO2017 datasets, respectively, compared to the baseline (YOLOV7-Tiny-SiLu), outperforming current one-stage detectors.

Owing to the lack of sufficient pixels and details, small object detection has long been a challenging and significant task in the field of detection. The traditional anchor-free detectors use vanilla sampling methods and only utilize very little regional information. This may lose a lot of information and can not utilize contextual information due to the relatively small receptive field. Therefore, we propose SCMBox to solve these problems. Compared to the current anchor-free techniques, our anchor-free approach includes several novel contributions. The entire network structure avoid using strided convolution layers and pooling layers, which makes the approach suitable for small objects with low resolution. The quality of feature upsampling directly influences the detection quality in small object detection. To preserve the global structure of the network as much as possible, we employed a highly effective upsampling operator that can make use of more contextual information. We proposed a modified receptive field block (MRFB) to increase the receptive field. We use Wasserstein distance in the loss function of the box regression to reduce the susceptibility of small objects to position variation. Our method achieves excellent performance on the standard MS-COCO object detection benchmark and Vis-Drone2021 dataset and outperforms the state-of-the-art methods.

The Internet of Medical Things (IoMT) is a brand new technology of combining medical devices and other wireless devices to access to the healthcare management systems. This article has sought the possibilities of aiding the current Corona Virus Disease 2019 (COVID-19) pandemic by implementing machine learning algorithms while offering emotional treatment suggestion to the doctors and patients. The cognitive model with respect to IoMT is best suited to this pandemic as every person is to be connected and monitored through a cognitive network. However, this COVID-19 pandemic still remain some challenges about emotional solicitude for infants and young children, elderly, and mentally ill persons during pandemic. Confronting these challenges, this article proposes an emotion-aware and intelligent IoMT system, which contains information sharing, information supervision, patients tracking, data gathering and analysis, healthcare, etc. Intelligent IoMT devices are connected to collect multimodal data of patients in a surveillance environments. The latest data and inputs from official websites and reports are tested for further investigation and analysis of the emotion analysis. The proposed novel IoMT platform enables remote health monitoring and decision-making about the emotion, therefore greatly contribute convenient and continuous emotion-aware healthcare services during COVID-19 pandemic. Experimental results on some emotion data indicate that the proposed framework achieves significant advantage when compared with the some mainstream models. The proposed cognition-based dynamic technology is an effective solution way for accommodating a big number of devices and this COVID-19 pandemic application. The controversy and future development trend are also discussed.

Automatic violence detection from video is a hot topic for many video surveillance applications. However, there has been little success in developing an algorithm that can detect violence in surveillance videos with high performance. In this paper, following our recently proposed idea of motion Weber local descriptor (WLD), we make two major improvements and propose a more effective and efficient algorithm for detecting violence from motion images. First, we propose an improved WLD (IWLD) to better depict low-level image appearance information, and then extend the spatial descriptor IWLD by adding a temporal component to capture local motion information and hence form the motion IWLD. Second, we propose a modified sparse-representation-based classification model to both control the reconstruction error of coding coefficients and minimize the classification error. Based on the proposed sparse model, a class-specific dictionary containing dictionary atoms corresponding to the class labels is learned using class labels of training samples. With this learned dictionary, not only the representation residual but also the representation coefficients become discriminative. A classification scheme integrating the modified sparse model is developed to exploit such discriminative information. The experimental results on three benchmark data sets have demonstrated the superior performance of the proposed approach over the state of the arts.

In this paper, we address the problem of human action recognition by representing image sequences as a sparse collection of patch-level spatiotemporal events that are salient in both space and time domain. Our method uses a multi-scale volumetric representation of video and adaptively selects an optimal space-time scale under which the saliency of a patch is most significant. The input image sequences are first partitioned into non-overlapping patches. Then, each patch is represented by a vector of coefficients that can linearly reconstruct the patch from a learned dictionary of basis patches. We propose to measure the spatiotemporal saliency of patches using Shannon's self-information entropy, where a patch's saliency is determined by information variation in the contents of the patch's spatiotemporal neighborhood. Experimental results on two benchmark datasets demonstrate the effectiveness of our proposed method.

Road sign detection plays an important role in driver assistance system. However, it faces problems of high computational cost and low contrast in video sequences. In this paper, we propose a two-level hierarchical algorithm which addresses these problems by making better use of the color and shape information of road signs. In order to solve the problem of low image contrast, we propose to improve the color contrast using our algorithm based on visual saliency. In order to reduce the high computational cost, an improved radial symmetry transform (IRST) is developed for grouping feature points on the basis of their underlying symmetry in an image. Experimental results show that our methods are robust to a broad range of lighting conditions and efficient enough for real-time applications.

In this paper, we propose a novel approach for violence detection and localization in a public scene. Currently, violence detection is considerably under-researched compared with the common action recognition. Although existing methods can detect the presence of violence in a video, they cannot precisely locate the regions in the scene where violence is happening. This paper will tackle the challenge and propose a novel method to locate the violence location in the scene, which is important for public surveillance. The Gaussian Mixed Model is extended into the optical flow domain in order to detect candidate violence regions. In each region, a new descriptor, Histogram of Optical Flow Orientation (HOFO), is proposed to measure the spatial-temporal features. A linear SVM is trained based on the descriptor. The performance of the method is demonstrated on the publicly available data sets, BEHAVE and CAVIAR.