Search

Hyperspectral image (HSI) and synthetic aperture radar (SAR) data joint classification is a crucial and yet challenging task in the field of remote sensing image interpretation. However, feature modeling in existing methods is deficient to exploit the abundant global, spectral, and local features simultaneously, leading to sub-optimal classification performance. To solve the problem, we propose a hierarchical attention and parallel filter fusion network for multi-source data classification. Concretely, we design a hierarchical attention module for hyperspectral feature extraction. This module integrates global, spectral, and local features simultaneously to provide more comprehensive feature representation. In addition, we develop parallel filter fusion module which enhances cross-modal feature interactions among different spatial locations in the frequency domain. Extensive experiments on two multi-source remote sensing data classification datasets verify the superiority of our proposed method over current state-of-the-art classification approaches. Specifically, our proposed method achieves 91.44% and 80.51% of overall accuracy (OA) on the respective datasets, highlighting its superior performance., Comment: Accepted by IEEE GRSL

Cross-Domain Few-Shot Learning has witnessed great stride with the development of meta-learning. However, most existing methods pay more attention to learning domain-adaptive inductive bias (meta-knowledge) through feature-wise manipulation or task diversity improvement while neglecting the phenomenon that deep networks tend to rely more on high-frequency cues to make the classification decision, which thus degenerates the robustness of learned inductive bias since high-frequency information is vulnerable and easy to be disturbed by noisy information. Hence in this paper, we make one of the first attempts to propose a Frequency-Aware Prompting method with mutual attention for Cross-Domain Few-Shot classification, which can let networks simulate the human visual perception of selecting different frequency cues when facing new recognition tasks. Specifically, a frequency-aware prompting mechanism is first proposed, in which high-frequency components of the decomposed source image are switched either with normal distribution sampling or zeroing to get frequency-aware augment samples. Then, a mutual attention module is designed to learn generalizable inductive bias under CD-FSL settings. More importantly, the proposed method is a plug-and-play module that can be directly applied to most off-the-shelf CD-FLS methods. Experimental results on CD-FSL benchmarks demonstrate the effectiveness of our proposed method as well as robustly improve the performance of existing CD-FLS methods. Resources at https://github.com/tinkez/FAP_CDFSC.

Multi-source remote sensing data classification has emerged as a prominent research topic with the advancement of various sensors. Existing multi-source data classification methods are susceptible to irrelevant information interference during multi-source feature extraction and fusion. To solve this issue, we propose a sparse focus network for multi-source data classification. Sparse attention is employed in Transformer block for HSI and SAR/LiDAR feature extraction, thereby the most useful self-attention values are maintained for better feature aggregation. Furthermore, cross-attention is used to enhance multi-source feature interactions, and further improves the efficiency of cross-modal feature fusion. Experimental results on the Berlin and Houston2018 datasets highlight the effectiveness of SF-Net, outperforming existing state-of-the-art methods., Comment: Accepted by IEEE IGARSS 2024

Hyperspectral image (HSI) contains abundant spatial and spectral information, making it highly valuable for unmixing. In this paper, we propose a Dual-Stream Attention Network (DSANet) for HSI unmixing. The endmembers and abundance of a pixel in HSI have high correlations with its adjacent pixels. Therefore, we adopt a "many to one" strategy to estimate the abundance of the central pixel. In addition, we adopt multiview spectral method, dividing spectral bands into multiple partitions with low correlations to estimate abundances. To aggregate the estimated abundances for complementary from the two branches, we design a cross-fusion attention network to enhance valuable information. Extensive experiments have been conducted on two real datasets, which demonstrate the effectiveness of our DSANet., Comment: Accepted by IEEE IGARSS 2024

Predicting accurate normal maps of objects from two-dimensional images in regions of complex structure and spatial material variations is challenging using photometric stereo methods due to the influence of surface reflection properties caused by variations in object geometry and surface materials. To address this issue, we propose a photometric stereo network called a RMAFF-PSN that uses residual multiscale attentional feature fusion to handle the ``difficult'' regions of the object. Unlike previous approaches that only use stacked convolutional layers to extract deep features from the input image, our method integrates feature information from different resolution stages and scales of the image. This approach preserves more physical information, such as texture and geometry of the object in complex regions, through shallow-deep stage feature extraction, double branching enhancement, and attention optimization. To test the network structure under real-world conditions, we propose a new real dataset called Simple PS data, which contains multiple objects with varying structures and materials. Experimental results on a publicly available benchmark dataset demonstrate that our method outperforms most existing calibrated photometric stereo methods for the same number of input images, especially in the case of highly non-convex object structures. Our method also obtains good results under sparse lighting conditions., Comment: 17 pages,12 figures

We present a complete study of the $B\to D^*P\ell^+\nu_\ell~(\ell=e,\mu,\tau)$ decays with the non-resonant, the charmed axial vector resonant and the charmed tensor resonant contributions by using the SU(3) flavor symmetry. Relevant amplitude relations between different decay modes are obtained by the SU(3) flavor symmetry. We then predict non-measured branching ratios of the $B\to D^*P\ell^+\nu_\ell$ decays with the non-resonant and the charmed resonant contributions by using present experimental data of the $B\to D^*P\ell'^+\nu_{\ell'}~(\ell'=e,\mu)$ decays within $2\sigma$ errors. We have found that $B^{0,+}\to D^*\eta\ell^+\nu_\ell$, $B^{0,+}\to D^*\eta'\ell^+\nu_\ell$, $B^{0}_s\to D^*_s \eta\ell^+\nu_\ell$, $B^{0}_s\to D^*_s\eta'\ell^+\nu_\ell$ and $B^{0,+}\to D^{*}_sK\ell^+\nu_\ell$ decays only receive non-resonant contributions. Decays $B^0_s\to D_s^{*-}\pi^0\ell^+\nu_\ell$ only receive the $D'_{s1}$ resonant contributions. Other decays receive all three kinds of contributions, and three kinds of contributions are important in most of decays., Comment: 15 pages, 1 figure

In the present work, the cluster radioactivity preformation probability Pc in the scheme of NpNn for the effective number of the valence particles (holes) in trans-lead nuclei has been systematically investigated. This quantity has been explored in the simplified parametrization of NpNn as well as the multiplication NpNnI of this product with the isospin asymmetry I. The calculations for Pc are both performed in microscopic and model-dependent way. Within the microscopic approach, based on our previous work [Chin. Phys. C 47,014101 (2023)], Pc is calculated in cluster formation model (CFM) combined with the exponential relationship of Pc to the alpha decay preformation probability P alpha when the mass number of the emitted cluster Ac less than 28. While Ac greater than 28, Pc is obtained through the charge-number dependence of Pc on the decay products proposed by Ren et al. [Phys. Rev. C 70,034304 (2004)]. In the model-dependent approach, Pc is extracted through the ratios from calculated cluster radioactivity half-lives in the framework of unified fission model (UFM) proposed by Dong et al. [Eur. Phys. J. A 41,197 (2009)] to experimental ones. Both of the results show Pc in logarithmic form are linear to NpNn as well as NpNnI. For comparison, the parent-mass-number dependence analytical formula as well as the model proposed by K. Wei and H. F. Zhang [Phys. Rev. C 96,021601(R)(2017)] are also used. Furthermore, the preformation mechanic for cluster radioactivity has also been discussed.

In the present study, proton emission half-lives have been investigated for the deformed proton emitters with $53\leq Z \leq 83$ in the deformed Gamow-like model, where the deformation effect has been included in the Coulomb potential. The experimental half-lives of proton emitters can be reproduced within a factor of 3.45. For comparison, other results from the universal decay law and the new Geiger-Nuttall law are presented as well. Furthermore, the relevance of the half-lives to the angular momentum $l$ for $^{117}$La, $^{121}$Pr, $^{135}$Tb and $^{141}$Ho has been analyzed, and corresponding possible values of $l$ has been put forward: $l=$3, 3, 4, 4.

In the present work, we systematically study the spectroscopic factor of proton radioactivity ($S_p$) with $A>100$ using the deformed two-potential approach (D-TPA). It is found that there is a link between the quadrupole deformation parameter of proton emitter and $S_p$. Based on this result, we propose a simple analytic formula for estimating the spectroscopic factor of proton radioactivity. With the help of this formula, the calculated half-lives of proton radioactivity can reproduce the experimental data successfully within a factor of 2.77. Furthermore, we extend the D-TPA with this formula for evaluating the spectroscopic factor to predict the proton radioactivity half-lives of 12 proton radioactivity candidates whose radioactivity is energetically allowed or observed but not yet quantified in NUBASE2020. For comparison, the universal decay law for proton radioactivity (UDLP) and the new Geiger-Nuttall law (NG-N) are also used. It turns out that all of the predicted results are basically consistent with each other.

We study the single parton fragmentation functions (FFs) at the input factorization scale $\mu_0\gtrsim 2m_Q$, with heavy quark mass $m_Q$, in the soft gluon factorization (SGF) approach. We express the FFs in terms of perturbatively calculable short distance hard parts for producing a heavy quark-antiquark pair in all possible states, convoluted with corresponding soft gluon distribution for the hadronization of the pair to a heavy quarkonium. We compute the perturbative short distance hard parts for producing a heavy quark pair in all possible $S$-wave and $P$-wave states up to $O(\alpha_s^2)$. With our results, the SGF can be further used to study the heavy quarkonium production at the hadron colliders and heavy quarkonium production within a jet., Comment: 19 pages, 0 figures. arXiv admin note: text overlap with arXiv:2304.04552

Controlled remote implementation of operation (CRIO) enables to implement operations on a remote state with strong security. We transmit implementations by entangling qubits in photon-cavity-atom system. The photons transferring in fibre and the atoms embedded in optical cavity construct CZ gates. The gates transfer implementations between participants with the permission of controller. We also construct nonadiabatic holonomic controlled gate between alkali metal atoms. Decoherence and dissipation decrease the fidelity of the implementation operators. We apply anti-blockade effect and dynamical scheme to improve the robustness of the gate.

Mining the hierarchical structure of Internet review topics and realizing a fine classification of review texts can help alleviate users' information overload. However, existing hierarchical topic classification methods primarily rely on external corpora and human intervention. This study proposes a Modified Continuous Renormalization (MCR) procedure that acts on the keyword co-occurrence network with fractal characteristics to achieve the topic hierarchy mining. First, the fractal characteristics in the keyword co-occurrence network of Internet review text are identified using a box-covering algorithm for the first time. Then, the MCR algorithm established on the edge adjacency entropy and the box distance is proposed to obtain the topic hierarchy in the keyword co-occurrence network. Verification data from the Dangdang.com book reviews shows that the MCR constructs topic hierarchies with greater coherence and independence than the HLDA and the Louvain algorithms. Finally, reliable review text classification is achieved using the MCR extended bottom level topic categories. The accuracy rate (P), recall rate (R) and F1 value of Internet review text classification obtained from the MCR-based topic hierarchy are significantly improved compared to four target text classification algorithms., Comment: 43 pages, 8 figures, conference or other essential info

The box-covering method plays a fundamental role in the fractal property recognition and renormalization analysis of complex networks. This study proposes the hub-collision avoidance and leaf-node options (HALO) algorithm. In the box sampling process, a forward sampling rule (for avoiding hub collisions) and a reverse sampling rule (for preferentially selecting leaf nodes) are determined for bidirectional network traversal to reduce the randomness of sampling. In the box selection process, the larger necessary boxes are preferentially selected to join the solution by continuously removing small boxes. The compact-box-burning (CBB) algorithm, the maximum-excluded-mass-burning (MEMB) algorithm, the overlapping-box-covering (OBCA) algorithm, and the algorithm for combining small-box-removal strategy and maximum box sampling with a sampling density of 30 (SM30) are compared with HALO in experiments. Results on nine real networks show that HALO achieves the highest performance score and obtains 11.40%, 7.67%, 2.18%, and 8.19% fewer boxes than the compared algorithms, respectively. The algorithm determinism is significantly improved. The fractal dimensions estimated by covering four standard networks are more accurate. Moreover, different from MEMB or OBCA, HALO is not affected by the tightness of the hubs and exhibits a stable performance in different networks. Finally, the time complexities of HALO and the compared algorithms are all O(N^2), which is reasonable and acceptable., Comment: 32 pages, 7 figures, conference or other essential info

In the present work, based on Wentzel-Kramers-Brillouin theory, we systematically study the cluster radioactivity half-lives of 22 nuclei ranging from $^{221}$$\rm{Fr}$ to $^{242}$$\rm{Cm}$ by using a phenomenological model, which considers the screened electrostatic effect of Coulomb potential. In this model, there are two adjustable parameters i.e. the parameter $t$ and $g$, which are related to the screened electrostatic barrier and the strength of spectroscopic factor, respectively. The calculated results indicate this model can well reproduce the experimental data while the corresponding root-mean-square (rms) deviation is 0.660. In addition, we extend this model to predict the half-lives of possible cluster radioactive candidates whose cluster radioactivity are energetically allowed or observed but not yet quantified in the evaluated nuclear properties table NUBASE2020. The predicted results are consistent with the ones obtained by using other theoretical models and/or empirical formulae including the universal decay law (UDL) proposed by Qi \textit{et al.} [Phys. Rev. C 80, 044326 (2009)], a semi-empirical model for both $\alpha$ decay and cluster radioactivity proposed by Santhosh \textit{et al.} [J. Phys. G 35, 085102 (2008)] and a unified formula of half-lives for $\alpha$ decay and cluster radioactivity proposed by Ni \textit{et al.} [Phys. Rev. C 78, 044310 (2008)].

Contrastive Language-Image Pre-training (CLIP) starts to emerge in many computer vision tasks and has achieved promising performance. However, it remains underexplored whether CLIP can be generalized to 3D hand pose estimation, as bridging text prompts with pose-aware features presents significant challenges due to the discrete nature of joint positions in 3D space. In this paper, we make one of the first attempts to propose a novel 3D hand pose estimator from monocular images, dubbed as CLIP-Hand3D, which successfully bridges the gap between text prompts and irregular detailed pose distribution. In particular, the distribution order of hand joints in various 3D space directions is derived from pose labels, forming corresponding text prompts that are subsequently encoded into text representations. Simultaneously, 21 hand joints in the 3D space are retrieved, and their spatial distribution (in x, y, and z axes) is encoded to form pose-aware features. Subsequently, we maximize semantic consistency for a pair of pose-text features following a CLIP-based contrastive learning paradigm. Furthermore, a coarse-to-fine mesh regressor is designed, which is capable of effectively querying joint-aware cues from the feature pyramid. Extensive experiments on several public hand benchmarks show that the proposed model attains a significantly faster inference speed while achieving state-of-the-art performance compared to methods utilizing the similar scale backbone., Comment: Accepted In Proceedings of the 31st ACM International Conference on Multimedia (MM' 23)

The task of Camouflaged Object Detection (COD) aims to accurately segment camouflaged objects that integrated into the environment, which is more challenging than ordinary detection as the texture between the target and background is visually indistinguishable. In this paper, we proposed a novel Feature Grafting and Distractor Aware network (FDNet) to handle the COD task. Specifically, we use CNN and Transformer to encode multi-scale images in parallel. In order to better explore the advantages of the two encoders, we design a cross-attention-based Feature Grafting Module to graft features extracted from Transformer branch into CNN branch, after which the features are aggregated in the Feature Fusion Module. A Distractor Aware Module is designed to explicitly model the two possible distractors in the COD task to refine the coarse camouflage map. We also proposed the largest artificial camouflaged object dataset which contains 2000 images with annotations, named ACOD2K. We conducted extensive experiments on four widely used benchmark datasets and the ACOD2K dataset. The results show that our method significantly outperforms other state-of-the-art methods. The code and the ACOD2K will be available at https://github.com/syxvision/FDNet., Comment: ICME2023 paper

Existing edge-aware camouflaged object detection (COD) methods normally output the edge prediction in the early stage. However, edges are important and fundamental factors in the following segmentation task. Due to the high visual similarity between camouflaged targets and the surroundings, edge prior predicted in early stage usually introduces erroneous foreground-background and contaminates features for segmentation. To tackle this problem, we propose a novel Edge-aware Mirror Network (EAMNet), which models edge detection and camouflaged object segmentation as a cross refinement process. More specifically, EAMNet has a two-branch architecture, where a segmentation-induced edge aggregation module and an edge-induced integrity aggregation module are designed to cross-guide the segmentation branch and edge detection branch. A guided-residual channel attention module which leverages the residual connection and gated convolution finally better extracts structural details from low-level features. Quantitative and qualitative experiment results show that EAMNet outperforms existing cutting-edge baselines on three widely used COD datasets. Codes are available at https://github.com/sdy1999/EAMNet., Comment: ICME2023 paper