The infant cry recognition methods predominantly employ a single audio feature, which makes it difficult to characterize the semantic information of cry signals in a comprehensive manner. The few studies that employ multiple features often adopt simple concatenation-based early fusion strategies, which lack deep cross-feature interaction. Meanwhile, public datasets are small in scale and limit improvements in recognition accuracy. To address these issues, this paper proposes a dual-branch intermediate feature fusion model. The model consists of a Mel-spectrogram branch based on Audio Spectrogram Transformer(AST) and an MFCC branch based on CNN-LSTM-Attention network. Global spectro-temporal patterns are extracted from mel-spectrograms by the AST encoder, while refined core acoustic features are captured by the convolutional-recurrent architecture with attention-based pooling in the MFCC branch. Features from both branches are fused through a cross-attention module to better explore and utilize the multi-level information in cry signals. Experimental results on the Donate_Cry and Baby_Crying datasets indicate that this method achieves accuracies of 86.8% and 58.8%, respectively, and outperforms the existing methods.

To promote the practicality of deep learning-based localization, existing studies aim to address the issue of scenario dependence through meta-learning. However, these studies primarily focus on variations in environmental layouts while overlooking the impact of changes in device configurations, such as bandwidth, the number of access points (APs), and the number of antennas used. Unlike environmental changes, variations in device configurations affect the dimensionality of channel state information (CSI), thereby compromising neural network usability. To address this issue, we propose Meta-SimGNN, a novelWiFi localization system that integrates graph neural networks with meta-learning to improve localization generalization and robustness. First, we introduce a fine-grained CSI graph construction scheme, where each AP is treated as a graph node, allowing for adaptability to changes in the number of APs. To structure the features of each node, we propose an amplitude-phase fusion method and a feature extraction method. The former utilizes both amplitude and phase to construct CSI images, enhancing data reliability, while the latter extracts dimension-consistent features to address variations in bandwidth and the number of antennas. Second, a similarity-guided meta-learning strategy is developed to enhance adaptability in diverse scenarios. The initial model parameters for the fine-tuning stage are determined by comparing the similarity between the new scenario and historical scenarios, facilitating rapid adaptation of the model to the new localization scenario. Extensive experimental results over commodity WiFi devices in different scenarios show that Meta-SimGNN outperforms the baseline methods in terms of localization generalization and accuracy.

WiFi-based indoor localization, supported by comprehensive infrastructure, is considered a highly promising solution. However, practical applications of existing WiFi-based methods often struggle due to dynamic antenna configurations and potential influence from obstacles, which can undermine the reliability of channel state information and frustrate localization. Even worse, environmental changes may lead to a domain shift, further degrading localization accuracy and system robustness. To address these problems, this paper introduces GraphFi, a novel system that leverages graph neural networks (GNNs) to deliver accurate and robust localization using nearby access points (APs). GraphFi designs two types of graph structures: intra-AP graph and inter-AP graph, to maximize the use of information from all available APs. They aggregate the local features among antennas within each AP and global features across APs, effectively addressing the problem of dynamic antenna configurations. Two specialized GNNs are utilized to derive accurate user locations from these graphs. Additionally, we introduce an anomaly detection method to identify and exclude obstacle-affected APs. This method also employs a tailored GNN to mitigate influence from unpredictable obstacles. Furthermore, we integrate an unsupervised domain adaptation mechanism based on a gradient reversal layer into GNNs. This helps maintain localization performance in a cost-efficient manner and ensures sustained effectiveness in a cross-domain setting. We prototype GraphFi using commodity WiFi devices and conduct extensive experiments in various scenarios. The results demonstrate that GraphFi achieves average localization errors of 0.17 m in a single-domain setting and 0.2851 m in a cross-domain setting, surpassing existing solutions in both precision and robustness.

Recent deep learning-based localization methods leverage meta-learning to enhance adaptability across diverse environments. However, most existing approaches focus on variations in environmental layouts while overlooking the changes in device configurations—such as the number of access points, antennas, or bandwidth. Unlike environmental variations, changes in device configurations fundamentally alter the dimensionality of channel state information (CSI), which can significantly hinder the usability and generalizability of neural networks. To address this problem, we propose Meta-GLoc, an adaptive and robust WiFi localization system that combines meta-learning with graph neural networks to effectively handle variations in CSI dimensionality. Specifically, we introduce an amplitude-phase fusion method and a feature extraction method to construct fine-grained CSI graphs. The former fuses the cleaned amplitude and phase in a carefully determined ratio to construct robust CSI images, while the latter extracts dimension-consistent features to mitigate the impact of varying bandwidth and antenna configurations. Moreover, meta-learning is employed to realize adaptive localization in different environments. Experiment results on commodity WiFi devices across different configurations demonstrate that Meta-GLoc effectively improves localization accuracy and robustness.

We propose Network Automatic Relevance Determination (NARD), an extension of ARD for linearly probabilistic models, to simultaneously model sparse relationships between inputs $X \in \mathbb R^{d \times N}$ and outputs $Y \in \mathbb R^{m \times N}$, while capturing the correlation structure among the $Y$. NARD employs a matrix normal prior which contains a sparsity-inducing parameter to identify and discard irrelevant features, thereby promoting sparsity in the model. Algorithmically, it iteratively updates both the precision matrix and the relationship between $Y$ and the refined inputs. To mitigate the computational inefficiencies of the $\mathcal O(m^3 + d^3)$ cost per iteration, we introduce Sequential NARD, which evaluates features sequentially, and a Surrogate Function Method, leveraging an efficient approximation of the marginal likelihood and simplifying the calculation of determinant and inverse of an intermediate matrix. Combining the Sequential update with the Surrogate Function method further reduces computational costs. The computational complexity per iteration for these three methods is reduced to $\mathcal O(m^3+p^3)$, $\mathcal O(m^3 + d^2)$, $\mathcal O(m^3+p^2)$ respectively, where $p \ll d$ is the final number of features in the model. Our methods demonstrate significant improvements in computational efficiency with comparable performance on both synthetic and real-world datasets.

Synthetic aperture radar (SAR) images are widely utilized for the detection and recognition of aircraft targets. Unlike optical images, SAR images possess the advantage of being applicable in all weather conditions and at all times of the day. However, in contrast to optical images, aircraft typically appear as discrete points in SAR images, and their outlines are not distinctly clear. To effectively use the scatter information of aircraft in SAR images, a convolutional modulation scattering feature network (CMSF) is proposed in this paper. Firstly, a scattering feature extraction module is introduced to make full advantage of scattering information. Secondly, following convolution processing, the convolution modulation module is employed to generate a similar fraction matrix. Thirdly, the fusion of scattering features and convolution features is achieved through convolutional modulation and matrix multiplication. Finally, a four-stage convolutional processing is employed to recognize the aircraft target. Extensive experiments conducted on the SARAircraft-1.0 dataset demonstrate the effectiveness of the convolutional modulated scattering feature network for aircraft target classification in SAR images.

Detecting infrared targets in complex backgrounds is a demanding task, especially for small targets. Infrared small target is usually difficult to be detected accurately because of the complex background and the lack of rich color information and edge information. Based on this difficulty, a Multiscale Interactive Attention Network (MIA-Net) is proposed to accurately detect small infrared targets in complex backgrounds. Specifically, the feature skip interactive module (FSIM) is firstly designed for realizing multiscale feature information transmission and fusion, which solves the problem of losing infrared small target features after multiple convolutions. Based on this, a multiscale edge reconstruction (MER) block is designed to learn the edge features of small targets and achieve optimal reconstruction of their edges. Extensive experiments were conducted on the SIRST dataset to evaluate the proposed MIA-Net and the results show that our MIA-Net performs better in terms of IoU, Pd, and Fa.

The problem of insufficient samples has been limiting the performance of intelligent interpretation in Synthetic Aperture Radar (SAR) images. Humans have the ability to recognize new instances with only a few samples, indicating that attributes play a crucial role in recognition. Attributes can be shared across categories and provide a distinctive representation. Motivated by this fact, this paper proposes an attribute-guided network consisting of a base classifier (BC) and an attribute classifier (AC). Firstly, we design binary attributes for SAR objects to enable more distinct feature representations. Secondly, the images are mapped into a semantic embedding space by embedding the attribute vectors. Finally, the performance of few-shot classification in SAR images is improved by jointly optimizing the loss in both attribute space and deep feature space. Our attribute-based framework is validated through ablation experiments on the MSTAR dataset and a self-built SAR aircraft dataset.

The interpretation of colors in visualizations is facilitated when the assignments between colors and concepts in the visualizations match human's expectations, implying that the colors can be interpreted in a semantic manner. However, manually creating a dataset of suitable associations between colors and concepts for use in visualizations is costly, as such associations would have to be collected from humans for a large variety of concepts. To address the challenge of collecting this data, we introduce a method to extract color-concept associations automatically from a set of concept images. While the state-of-the-art method extracts associations from data with supervised learning, we developed a self-supervised method based on colorization that does not require the preparation of ground truth color-concept associations. Our key insight is that a set of images of a concept should be sufficient for learning color-concept associations, since humans also learn to associate colors to concepts mainly from past visual input. Thus, we propose to use an automatic colorization method to extract statistical models of the color-concept associations that appear in concept images. Specifically, we take a colorization model pre-trained on ImageNet and fine-tune it on the set of images associated with a given concept, to predict pixel-wise probability distributions in Lab color space for the images. Then, we convert the predicted probability distributions into color ratings for a given color library and aggregate them for all the images of a concept to obtain the final color-concept associations. We evaluate our method using four different evaluation metrics and via a user study. Experiments show that, although the state-of-the-art method based on supervised learning with user-provided ratings is more effective at capturing relative associations, our self-supervised method obtains overall better results according to metrics like Earth Mover's Distance (EMD) and Entropy Difference (ED), which are closer to human perception of color distributions.

Predictive energy management has become a new focus of the automobile industry for its high potential of further reducing energy consumption. Based on previous works on predictive speed optimization using discrete dynamic programming (DDP), this paper introduces a novel approach of applying DDP with variable step size in stage variable discretization, which can realize a better tradeoff between precision and computational cost. In this approach, a “meshing” algorithm searches the points of interest (POI), such as speed limit change, traffic lights, and road curvatures, where changes in vehicle speed are expected. The algorithm increases the step-size resolution close to these points and reduces the resolutions in positions further away from POI, where the optimized vehicle speed is insensitive to the step size. With this approach, the position of POI can be precisely located to solve the DDP problem. In a test case with a relatively high density of POI, the computational cost is reduced by more than 53% by only sacrificing less than 1% of precision compared to a fixed step-size discretization with high resolutions. It can be expected that, with a lower density of POI, the computational cost will be reduced even further.