In dynamic systems, complex time-varying noise characteristics impose challenges for multi-sensor data fusion. This paper proposes a deep learning-driven noise-adaptive filtering approach for multi-sensor data fusion. A parallel gated recurrent unit (GRU) and one-dimensional convolutional neural network (1D-CNN) architecture jointly extracts temporal dependencies and local spatial patterns from raw measurement sequences to estimate sensor noise variances. Each sensor then uses these variance estimates for more accurate adaptive local filtering. Based on these refined local estimates, the fusion center employs an event-triggered scheme to drastically cut communications while preserving fusion accuracy. Comprehensive experiments in a high-fidelity Unreal Engine–AirSim quadrotor unmanned aerial vehicle (UAV) simulation under nominal, drift, abrupt, and extreme noise modes confirm that our pipeline achieves superior global state estimation while dramatically reducing data transmission burden.

Current parameterized image representations embed visual information along the semantic boundaries and struggle to express the internal detailed texture structures of image components, leading to a lack of content consistency after image editing and driving. To address these challenges, this work proposes a novel parameterized representation based on hierarchical image proxy geometry, utilizing multi-layer hierarchically interrelated proxy geometric control points to embed multi-scale long-range structures and fine-grained texture details. The proposed representation enables smoother and more continuous interpolation during image rendering and ensures high-quality consistency within image components during image editing. Additionally, under the layer-wise representation strategy based on semantic-aware image layer decomposition, we enable decoupled image shape/texture editing of the targets of interest within the image. Extensive experimental results on image vectorization and editing tasks demonstrate that our proposed method achieves high rendering accuracy of general images, including natural images, with a significantly higher image parameter compression ratio, facilitating user-friendly editing of image semantic components.

With the advent of the Industry 4.0 era, the widespread deployment of Automated Guided Vehicles (AGVs) in factories has enabled them to serve as sensor relays, assisting in collecting sensor data in areas with poor signal quality. Traditionally, the objective of sensor data collection has been primarily to reduce the delay in data acquisition. However, latency alone offers an incomplete reflection of the significance of sensor data to industrial tasks. Value of information (VoI) has emerged as a novel metric that more accurately reflects the impact of sensor data on the performance of upstream tasks. In this background, we introduce an innovative AGV-assisted sensor data collection strategy to minimize the loss of sensor data VoI. This strategy encompasses the selection of data fusion nodes, choice of transmission modes, and AGV path planning. We introduce a new metric called structural value entropy, which effectively reduces VoI loss during the data fusion process, and through the design of a metaheuristic algorithm based on ant colony optimization, achieves the selection of transmission modes and the planning of AGV paths with minimal VoI loss. Simulation experiments validate the effectiveness of the proposed strategy in maintaining VoI, demonstrating significant performance enhancements and acceptable convergence speed compared to baseline strategies, affirming the strategy's efficiency and feasibility in handling large-scale sensor data collection tasks.

The development of speech synthesis technology has increased the attention toward the threat of spoofed speech. Although various high-performance spoofing countermeasures have been proposed in recent years, a particular scenario is overlooked: partially spoofed audio, where spoofed utterances may contain both spoofed and bona fide segments. Currently, the research on partially spoofed speech detection is lacking. The existing methods either train with partially spoofed speech at utterance level, resulting in gradient conflicting at the segment level, or directly train with segment level data, which requires segment labels that are difficult to obtain in practice. In this study, to better detect partially spoofed speech when only utterance labels are available, we formulate partially spoofed speech detection into a multiple instance learning (MIL) problem. The typical MIL uses a pooling layer to fuse patch scores as a whole, and we propose a hybrid MIL (H-MIL) framework based on max and log-sum-exp pooling methods, which can learn better segment representations to improve partially spoofed speech detection performance. Theoretical and experimental verification shows that H-MIL can effectively relieve the gradient conflicting and gradient vanishing problems. In addition, we analyze the local correlations between segments and introduce a local self-attention mechanism to enhance segment features, which further promotes the detection performance. In our experiments, we provide not only detection results at the segment and utterance levels but also some detailed visualization analysis, including the effect of spoof ratio and cross-dataset detection. The experimental results demonstrate the effective detection performance of our method at both the utterance and segment levels, especially when dealing with low spoof ratio attacks. The results confirm that our approach can better deal with partially spoofed speech detection than previous methods.

Owing to the widespread deployment of face and speaker recognition systems, research on attacks on neural-network-based biometric systems, which involves face or voice signal classification problems with a low-dimensional output vector, has drawn increasing attention. Recently, cross-modal voice-to-face (VTF) systems have learned to generate faces from voices by matching several biometric characteristics of the generated faces to those of speakers. However, attacks focusing on VTF systems with high-dimensional face image outputs have not yet been conducted. In this paper, we introduce various adversarial attack methods for the VTF system under different attack conditions. These methods can generate a fake face close to the target face or far from the original face, by adding subtle perturbations to the original voice. Under the white-box setting, we formulate a multiobjective optimization to generate target faces and improve the imperceptibility of the adversarial sample. Further a stepwise iterative optimization strategy is proposed to achieve faster and more effective attacks. Finally, the results of comparative experiments with various methods are demonstrated. Under the black-box setting, the adversarial samples generated from surrogate models are able to generate the fake face far from the original one. Qualitative and quantitative experimental results show the high target face-matching rate and irrelevance to the original face, as well as the imperceptibility of the adversarial audio. This study provides useful insights for privacy protection and improving generation robustness for information security.