Recent advancements in Visual Language Models (VLMs) have significantly driven research in open-vocabulary 3D scene reconstruction, showcasing strong potential in open-set retrieval and semantic understanding. However, existing approaches face challenges in open-world environments: they either suffer from insufficient precision in semantic segmentation, leading to inadequate fine-grained scene understanding, or they are limited to object-level reconstruction, failing to capture intricate object details and lack applicability in open-world settings. To address these issues, we introduce LE-Object, an object-centric Neural Implicit Radiance Field (NeRF) method for open-world scenarios to achieve fine-grained scene understanding and high-fidelity object reconstruction. LE-Object integrates spatial features (SF) from object point clouds with visual features (VF) from VLMs to perform object association, ensuring spatiotemporal consistency in object mask segmentation, and extends VLM features from 2D images into 3D space, enabling precise open-world semantic inference and detailed object reconstruction. Experimental results demonstrate that LE-Object excels in zero-shot semantic segmentation and open-world object reconstruction, offering innovative solutions for global navigation and local object manipulation in open-world applications.

In recent years, cross-domain few-shot learning has become a research hotspot in image classification tasks due to the diversity of classification scenarios in practical applications. Studies have shown that few-shot learning models are prone to overfitting during training, and domain distribution discrepancies exist between the source and target domains. These discrepancies make the feature extractor trained on the source domain poorly suited for the target domain. This paper proposes a Domain-Aligned Prototypical Network (DAPN) to alleviate domain distribution discrepancies. First, a small amount of unlabeled target domain data is collected and used for adversarial training with source domain data to promote distribution alignment between the source and target domains. Second, features at different levels are extracted during feature extraction, further enhancing cross-domain generalization capabilities. Finally, continuous optimization of the loss function enables the training of a feature extractor better suited to the target domain, thereby improving the model’s classification accuracy. Sufficient comparative experiments are conducted on eight commonly used cross-domain small-sample learning datasets, and the experimental results validate the effectiveness of the method. Code is available in https://github.com/yuanfangleshi/DAPN.git.

Recently, training deep neural networks (DNNs) on edge devices has attracted much attention due to its strong adaptability and avoidance of private data transmission. However, limited computational, storage, and energy resources pose significant challenges for edge devices. The structural and computational redundancies in DNNs create opportunities for sparse training through model pruning and zero-computation skipping. Although feasible, the sparse training accelerator design encounters common issues, such as redundant data duplication and unbalanced workloads, caused by irregular sparsity. To address these issues, this paper proposes a sparse DNN training accelerator, SDTA, together with a hierarchical pre-fetching buffer and a dynamic scheduler to achieve high design efficiency. The SDTA is deployed on the FPGA XCVU3P platform. Compared to the prior FPGA-based accelerators and the GPU, SDTA improves the energy efficiency by up to 2.29×, the storage utilization efficiency by up to 7.37×, and the computational efficiency by up to 1.9×. Compared to the dense accelerator, it achieves a speedup of up to 5.88×, while ensuring model accuracy.

Recent advancements in neural implicit fields for Simultaneous Localization and Mapping (SLAM) have provided breakthroughs. However, the benefits of reconstruction results to the perception ability of robot are minimal. Therefore, we propose FI-SLAM, a dense semantic instance SLAM system based on neural implicit representation, which significantly aids robots in better understanding the scene. FI-SLAM employs a coordinate and plane joint encoding method, which reduces the difficulty of feature storage by flattening the feature space. Furthermore, to improve representation efficiency, we use the method of adjacent feature level linear interpolation to describe features. We propose a feature fusion (FF) method to merge the object features with the scene features. The fused feature vector enhances the reconstruction accuracy of the local scene while ensuring the global reconstruction effect. It has improved the global reconstruction effect of the scene and the accuracy of camera tracking. Numerous experiments on synthetic and real-world datasets demonstrate that our method can assure accurate tracking precision, high-fidelity reconstruction results, and complete semantic instance maps. In summary, the algorithm we proposed heavily augments the scene perception capabilities of robot.

Managing healthcare data is essential for everyone, but it can be challenging and time-consuming. Although several Web and mobile applications existed for personal medical record management, there were few home-based, family-oriented solutions. In this research, we proposed to design an easy-to-use solution for a family. We designed and implemented an innovative Home Medical Care System, MedCare4Home, by deploying a MERN stack Web application to a minicomputer, Raspberry Pi 4, and enabling a local network to connect different devices at home. The users can access the Web application using either a desktop, tablet, mobile phone, or simply a 7-inch LCD touchscreen. MedCare4Home not only allows users to keep track of appointments and organize healthcare documents, but it can also help set up medication reminders and provide a self-report interface for symptom collection for family members easily and securely in one place. The final developed system demonstrated that we achieved the original goal and beyond. We expect to extend our current hardware design and software development to support physiological data collection with external sensors in the future.

OBJECTIVE: Preterm birth is the leading cause of neonatal morbidity and mortality. Early identification of high-risk patients followed by medical interventions is essential to the prevention of preterm birth. Based on the relationship between uterine contraction and the fundamental electrical activities of muscles, we extracted effective features from EHG signals recorded from pregnant women, and use them to train classifiers with the purpose of providing high precision in classifying term and preterm pregnancies. METHODS: To characterize changes from irregularity to coherence of the uterine activity during the whole pregnancy, network representations of the original electrohysterogram (EHG) signals are established by applying the Horizontal Visibility Graph (HVG) algorithm, from which we extract network degree density and distribution, clustering coefficient and assortativity coefficient. Concerns on the interferences of different noise sources embedded in the EHG signal, we apply Short-Time Fourier Transform (STFT) to expand the original signal in the time-frequency domain. This allows a network representation and the extraction of related features on each frequency component. Feature selection algorithms are then used to filter out unrelated frequency components. We further apply the proposed feature extraction method to EHG signals available in the Term-Preterm EHG database (TPEHG), and use them to train classifiers. We adopt the Partition-Synthesis scheme which splits the original imbalanced dataset into two sets, and synthesizes artificial samples separately within each subset to solve the problem of dataset imbalance. RESULTS: The optimally selected network-based features, not only contribute to the identification of the essential frequency components of uterine activities related to preterm birth, but also to improved performance in classifying term/preterm pregnancies, i.e., the SVM (Support Vector Machine) classifier trained with the available samples in the TPEHG gives sensitivity, specificity, overall accuracy, and auc values as high as 0.89, 0.93, 0.91, and 0.97, respectively.

We proposed Image2Vec, an unsupervised algorithm that learns feature representations for variable-size pieces of Chinese character images.