On the Rate-Distortion Theory for Task-Specific Semantic Communication.

Semantic communication has attracted considerable interest due to its potential to support emerging human-centric services, such as holographic communications, extended reality (XR), and human-machine interactions. Different from traditional communication systems that focus on minimizing the symbol-level distortion (e.g.

Alternating Direction Unfolding With a Cross Spectral Attention Prior for Dual-Camera Compressive Hyperspectral Imaging.

Coded Aperture Snapshot Spectral Imaging (CASSI) multiplexes 3D Hyperspectral Images (HSIs) into a 2D sensor to capture dynamic spectral scenes, which, however, sacrifices the spatial information. Dual-Camera Compressive Hyperspectral Imaging (DCCHI) enhances CASSI by incorporating a Panchromatic (PAN) camera to compensate for the loss of spatial information in CASSI. However, the dual-camera structure of DCCHI disrupts the diagonal property of the product of the sensing matrix and its transpose, making it difficult to efficiently and accurately solve the data subproblem in closed-form and thereby hindering the application of model-based methods and Deep Unfolding Networks (DUNs) that rely on such a closed-form solution.

Integrated fiber-PIC synthetic aperture chip enables 2D high-resolution interferometric imaging.

The implementation of synthetic aperture interferometry via photonic integrated circuits (PICs) holds significant potential for miniaturized high-resolution imaging systems, particularly in space optics and precision metrology. However, its advancement remains constrained by two fundamental limitations: restricted baseline reconfigurability (conventionally fixed at <20 mm) and aliasing artifacts induced by quasi-uniform sampling patterns during two-dimensional image reconstruction. To overcome these challenges, we develop a hybrid fiber-PIC architecture enabling continuous baseline adjustment from 10 mm to 50 mm through dynamically tunable fiber delay lines, integrated with a wheel-optimized non-uniform radial sampling strategy designed to minimize spectral matrix mapping errors by prioritizing baseline angles intersecting frequency grid points, achieving 1.

Multi-scale feature extraction and aggregation network for electroencephalography classification in face photo-sketch recognition task.

Face photo-sketch recognition task plays a crucial role in forensic investigation, human visual perception, and facial biometrics applications. The substantial modality gap between photographs and sketches, compounded by the influence of the semantic gap, poses a formidable challenge to recognition tasks. This study aims to propose an effective electroencephalography (EEG)-based approach to bridge this gap.

Effect of the dry-wet cycle on the mechanical properties of Ili loess with different mica contents.

The mechanical properties of the loess in the Ili region of China deteriorate significantly when it is subjected to the dry-wet cycles. Attributed to the critical role played by the mica content for the mechanical deterioration of Ili loess, a series of laboratory tests, including the X-ray diffraction (XRD) tests, the triaxial compression tests, the scanning electron microscopy (SEM), and other methods, were carried out to investigate the macroscopic and microscopic properties of Ili loess under different dry-wet cycles (i.e.

Self-Supervised Learning of LiDAR 3D Point Clouds Via 2D-3D Neural Calibration.

This paper introduces a novel self-supervised learning framework for enhancing 3D perception in autonomous driving scenes. Specifically, our approach, namely NCLR, focuses on 2D-3D neural calibration, a novel pretext task that estimates the rigid pose aligning camera and LiDAR coordinate systems. First, we propose the learnable transformation alignment to bridge the domain gap between image and point cloud data, converting features into a unified representation space for effective comparison and matching.

Spatio-Temporal Progressive Attention Model for EEG Classification in Rapid Serial Visual Presentation Task.

As a type of multi-dimensional sequential data, the spatial and temporal dependencies of electroencephalogram (EEG) signals should be further investigated. Thus, in this paper, we propose a novel spatial-temporal progressive attention model (STPAM) to improve EEG classification in rapid serial visual presentation (RSVP) tasks. STPAM employs a progressive approach using three sequential spatial experts to learn brain region topology and mitigate interference from irrelevant areas.

Applying SSVEP BCI on Dynamic Background.

Brain-computer interfaces (BCIs) based on steady-state visual evoked potential (SSVEP) have attracted much attention due to their high efficiency and accuracy. The SSVEP paradigm and decoding methods have been extensively studied and achieved remarkable results. This study proposed two modulation methods for the SSVEP paradigm, namely color inversion modulation and brightness compression modulation.

Recognizing human-object interactions in videos with the supervision of natural language.

Existing models for recognizing human-object interaction (HOI) in videos mainly rely on visual information for reasoning and generally treat recognition tasks as traditional multi-classification problems, where labels are represented by numbers. This supervised learning method discards semantic information in the labels and ignores advanced semantic relationships between actual categories. In fact, natural language contains a wealth of linguistic knowledge that humans have distilled about human-object interaction, and the category text contains a large amount of semantic relationships between texts.

SDL: Shift-Sensitive Spatial-Spectral Disentangling Learning for Hyperspectral Image Unsupervised Domain Adaptation.

Unsupervised domain adaptation (UDA) techniques, extensively studied in hyperspectral image (HSI) classification, aim to use labeled source domain data and unlabeled target domain data to learn domain invariant features for cross-scene classification. Compared to natural images, numerous spectral bands of HSIs provide abundant semantic information, but they also increase the domain shift significantly. In most existing methods, both explicit alignment and implicit alignment simply align feature distribution, ignoring domain information in the spectrum.

Confined Palladium Nanoparticles within a Robust Hierarchical Porous Zr-Based Metal-Organic Framework for Aromatic Alcohol Oxidation.

The development of novel host-guest catalysts with nanoscale active sites provides new opportunities to gain highly efficient catalytic performance for the oxidation of aromatic alcohols into high-value compounds. In this work, we develop a green and facile synthesis method to incorporate Pd nanoparticles (Pd) into a robust hierarchical porous Zr-metal-organic framework (Zr-IPA) through in situ solvent-free synthesis and wet reduction techniques. The optimized Pd/Zr-IPA catalyst exhibits exceptional benzyl alcohol oxidation performance, selectively oxidizing 95.

A multi-day and high-quality EEG dataset for motor imagery brain-computer interface.

A key challenge in developing a robust electroencephalography (EEG)-based brain-computer interface (BCI) is obtaining reliable classification performance across multiple days. In particular, EEG-based motor imagery (MI) BCI faces large variability and low signal-to-noise ratio. To address these issues, collecting a large and reliable dataset is critical for learning of cross-session and cross-subject patterns while mitigating EEG signals inherent instability.

Failure mechanism of soft- hard- interbedded rock slopes in cold regions: Numerical simulation and theoretical analysis.

The soft- hard- interbedded rock slope in cold regions generally undergo the differential weathering due to the freeze-thaw effects, for which the irregular rock fractures increase the risk of geological disasters occurrence. To investigate the failure mechanism of the rock slope, both the numerical simulation and theoretical analysis were adopted in the present research. The structural integrity of the soft- hard- interbedded rock slope subjected to three different conditions (e.

Joint Spatial and Frequency Domain Learning for Lightweight Spectral Image Demosaicing.

Conventional spectral image demosaicing algorithms rely on pixels' spatial or spectral correlations for reconstruction. Due to the missing data in the multispectral filter array (MSFA), the estimation of spatial or spectral correlations is inaccurate, leading to poor reconstruction results, and these algorithms are time-consuming. Deep learning-based spectral image demosaicing methods directly learn the nonlinear mapping relationship between 2D spectral mosaic images and 3D multispectral images.

Local Uncertainty Energy Transfer for Active Domain Adaptation.

Active Domain Adaptation (ADA) improves knowledge transfer efficiency from the labeled source domain to the unlabeled target domain by selecting a few target sample labels. However, most existing active sampling methods ignore the local uncertainty of neighbors in the target domain,making it easier to pick out anomalous samples that are detrimental to the model. To address this problem, we present a new approach to active domain adaptation called Local Uncertainty Energy Transfer (LUET), which integrates active learning of local uncertainty confusion and energy transfer alignment constraints into a unified framework.

CrossEI: Boosting Motion-oriented Object Tracking with An Event Camera.

With the differential sensitivity and high time resolution, event cameras can record detailed motion clues, which form a complementary advantage with frame-based cameras to enhance the object tracking, especially in challenging dynamic scenes. However, how to better match heterogeneous event-image data and exploit rich complementary cues from them still remains an open issue. In this paper, we align event-image modalities by proposing a motion adaptive event sampling method, and we revisit the cross-complementarities of event-image data to design a bidirectional-enhanced fusion framework.

Learning Frame-Event Fusion for Motion Deblurring.

Motion deblurring is a highly ill-posed problem due to the significant loss of motion information in the blurring process. Complementary informative features from auxiliary sensors such as event cameras can be explored for guiding motion deblurring. The event camera can capture rich motion information asynchronously with microsecond accuracy.

Research on optical interferometric imaging with flexible control using optical fibers and PIC chip.

We propose a prototype called a flexible integrated resolution and efficient light-imaging-expanded synthetic system (FIREFLIES). This paper describes the design, manufacturing, and experimental demonstration of the proposed system. FIREFLIES enables interferometric imaging at approximately 1550 nm using a variable baseline sampling technique, in which the baseline-collected light field forms interference fringes that are captured by an on-chip photodetector.

Multi-Modality Multi-Attribute Contrastive Pre-Training for Image Aesthetics Computing.

In the Image Aesthetics Computing (IAC) field, most prior methods leveraged the off-the-shelf backbones pre-trained on the large-scale ImageNet database. While these pre-trained backbones have achieved notable success, they often overemphasize object-level semantics and fail to capture the high-level concepts of image aesthetics, which may only achieve suboptimal performances. To tackle this long-neglected problem, we propose a multi-modality multi-attribute contrastive pre-training framework, targeting at constructing an alternative to ImageNet-based pre-training for IAC.

Impact of COVID-19 restrictions liberalization on air quality: a case study of Chongqing, Southwest China.

To mitigate the societal impact of the COVID-19 pandemic, China implemented long-term restrictive measures. The sudden liberalization at the end of 2022 disrupted residents' daily routines, making it scientifically intriguing to explore its effect on air quality. Taking Chongqing City in Southwest China as an example, we examined the impact of restriction liberalization on air quality, identified potential sources of pollutants, simulated the effects of abrupt anthropogenic control relaxation using a Random Forest Model, and applied an optimized model to predict the post-liberalization pollutant concentrations.

Fast Window-Based Event Denoising With Spatiotemporal Correlation Enhancement.

Previous deep learning-based event denoising methods mostly suffer from poor interpretability and difficulty in real-time processing due to their complex architecture designs. In this paper, we propose window-based event denoising, which simultaneously deals with a stack of events while existing element-based denoising focuses on one event each time. Besides, we give the theoretical analysis based on probability distributions in both temporal and spatial domains to improve interpretability.

Multi-Scale Spatio-Temporal Memory Network for Lightweight Video Denoising.

Deep learning-based video denoising methods have achieved great performance improvements in recent years. However, the expensive computational cost arising from sophisticated network design has severely limited their applications in real-world scenarios. To address this practical weakness, we propose a multiscale spatio-temporal memory network for fast video denoising, named MSTMN, aiming at striking an improved trade-off between cost and performance.

One-Pot Preparation of Nitro-Functionalized Bimetallic UiO-66(Zr-Hf) with Hierarchical Porosity for Oxidative Desulfurization Performance.

Preparation of multifunctional metal-organic frameworks (MOFs) offers new opportunities to obtain ultrahigh synergistic catalytic performance for heterogeneous reactions; however, the application of a one-pot method for preparing multifunctional MOFs remains challenging. Herein, we develop a one-pot green route for synthesizing bimetallic nitro-functionalized UiO-66(Zr-Hf)-NO with hierarchical porosity under solvent-free conditions. The optimal UiO-66(Zr-Hf)-NO shows an ultrahigh enhancement of oxidative desulfurization (ODS) efficiency to oxidize sulfur compounds (1000 ppm sulfur) in a model fuel at 40 °C within 12 min due to the introduction of more active Hf sites in the nodes, the increased Lewis acidity of Zr/Hf-O nodes by the electron-withdrawing NO group, and the enhanced diffusion rates by the mesopores.