The unsupervised multi-modal image translation is an emerging domain of computer vision whose goal is to transform an image from the source domain into many diverse styles in the target domain.However,the multi-genera...The unsupervised multi-modal image translation is an emerging domain of computer vision whose goal is to transform an image from the source domain into many diverse styles in the target domain.However,the multi-generator mechanism is employed among the advanced approaches available to model different domain mappings,which results in inefficient training of neural networks and pattern collapse,leading to inefficient generation of image diversity.To address this issue,this paper introduces a multi-modal unsupervised image translation framework that uses a generator to perform multi-modal image translation.Specifically,firstly,the domain code is introduced in this paper to explicitly control the different generation tasks.Secondly,this paper brings in the squeeze-and-excitation(SE)mechanism and feature attention(FA)module.Finally,the model integrates multiple optimization objectives to ensure efficient multi-modal translation.This paper performs qualitative and quantitative experiments on multiple non-paired benchmark image translation datasets while demonstrating the benefits of the proposed method over existing technologies.Overall,experimental results have shown that the proposed method is versatile and scalable.展开更多
The large volumetric variations experienced by metal selenides within conversion reaction result in inferior rate capability and cycling stability,ultimately hindering the achievement of superior electrochemical perfo...The large volumetric variations experienced by metal selenides within conversion reaction result in inferior rate capability and cycling stability,ultimately hindering the achievement of superior electrochemical performance.Herein,metallic Cu_(2)Se encapsulated with N-doped carbon(Cu_(2)Se@NC)was prepared using Cu_(2)O nanocubes as templates through a combination of dopamine polymerization and hightemperature selenization.The unique nanocubic structure and uniform N-doped carbon coating could shorten the ion transport distance,accelerate electron/charge diffusion,and suppress volume variation,ultimately ensuring Cu_(2)Se@NC with excellent electrochemical performance in sodium ion batteries(SIBs)and potassium ion batteries(PIBs).The composite exhibited excellent rate performance(187.7 mA h g^(-1)at 50 A g^(-1)in SIBs and 179.4 mA h g^(-1)at 5 A g^(-1)in PIBs)and cyclic stability(246,8 mA h g^(-1)at 10 A g^(-1)in SIBs over 2500 cycles).The reaction mechanism of intercalation combined with conversion in both SIBs and PIBs was disclosed by in situ X-ray diffraction(XRD)and ex situ transmission electron microscope(TEM).In particular,the final products in PIBs of K_(2)Se and K_(2)Se_(3)species were determined after discharging,which is different from that in SIBs with the final species of Na_(2)Se.The density functional theory calculation showed that carbon induces strong coupling and charge interactions with Cu_(2)Se,leading to the introduction of built-in electric field on heterojunction to improve electron mobility.Significantly,the theoretical calculations discovered that the underlying cause for the relatively superior rate capability in SIBs to that in PIBs is the agile Na~+diffusion with low energy barrier and moderate adsorption energy.These findings offer theoretical support for in-depth understanding of the performance differences of Cu-based materials in different ion storage systems.展开更多
针对红外图像纹理弱及多目标遮挡导致跟踪精度低的问题,构建了基于改进YOLOv7模型和多目标跟踪算法DeepSort的融合红外目标跟踪模型MSB-YOLOv7-DeepSort。采用SE(squeeze and excitation)通道注意力机制和双向特征金字塔网络提高红外目...针对红外图像纹理弱及多目标遮挡导致跟踪精度低的问题,构建了基于改进YOLOv7模型和多目标跟踪算法DeepSort的融合红外目标跟踪模型MSB-YOLOv7-DeepSort。采用SE(squeeze and excitation)通道注意力机制和双向特征金字塔网络提高红外目标的特征提取质量;利用轻量化网络MobileNetV3替换YOLOv7骨干网络,提升融合模型的推理速度。实验结果表明,MSB-YOLOv7-DeepSort模型在跟踪准确度、跟踪精确度、正确目标跟踪比例和帧率等方面均具有较好的性能。展开更多
基金the National Natural Science Foundation of China(No.61976080)the Academic Degrees&Graduate Education Reform Project of Henan Province(No.2021SJGLX195Y)+1 种基金the Teaching Reform Research and Practice Project of Henan Undergraduate Universities(No.2022SYJXLX008)the Key Project on Research and Practice of Henan University Graduate Education and Teaching Reform(No.YJSJG2023XJ006)。
文摘The unsupervised multi-modal image translation is an emerging domain of computer vision whose goal is to transform an image from the source domain into many diverse styles in the target domain.However,the multi-generator mechanism is employed among the advanced approaches available to model different domain mappings,which results in inefficient training of neural networks and pattern collapse,leading to inefficient generation of image diversity.To address this issue,this paper introduces a multi-modal unsupervised image translation framework that uses a generator to perform multi-modal image translation.Specifically,firstly,the domain code is introduced in this paper to explicitly control the different generation tasks.Secondly,this paper brings in the squeeze-and-excitation(SE)mechanism and feature attention(FA)module.Finally,the model integrates multiple optimization objectives to ensure efficient multi-modal translation.This paper performs qualitative and quantitative experiments on multiple non-paired benchmark image translation datasets while demonstrating the benefits of the proposed method over existing technologies.Overall,experimental results have shown that the proposed method is versatile and scalable.
基金The Natural Science Foundation of Henan Province(222300420083)the Opening Foundation of State Key Laboratory of Chemistry and Utilization of Carbon-based Energy Resource of Xinjiang University(KFKT2021004)。
文摘The large volumetric variations experienced by metal selenides within conversion reaction result in inferior rate capability and cycling stability,ultimately hindering the achievement of superior electrochemical performance.Herein,metallic Cu_(2)Se encapsulated with N-doped carbon(Cu_(2)Se@NC)was prepared using Cu_(2)O nanocubes as templates through a combination of dopamine polymerization and hightemperature selenization.The unique nanocubic structure and uniform N-doped carbon coating could shorten the ion transport distance,accelerate electron/charge diffusion,and suppress volume variation,ultimately ensuring Cu_(2)Se@NC with excellent electrochemical performance in sodium ion batteries(SIBs)and potassium ion batteries(PIBs).The composite exhibited excellent rate performance(187.7 mA h g^(-1)at 50 A g^(-1)in SIBs and 179.4 mA h g^(-1)at 5 A g^(-1)in PIBs)and cyclic stability(246,8 mA h g^(-1)at 10 A g^(-1)in SIBs over 2500 cycles).The reaction mechanism of intercalation combined with conversion in both SIBs and PIBs was disclosed by in situ X-ray diffraction(XRD)and ex situ transmission electron microscope(TEM).In particular,the final products in PIBs of K_(2)Se and K_(2)Se_(3)species were determined after discharging,which is different from that in SIBs with the final species of Na_(2)Se.The density functional theory calculation showed that carbon induces strong coupling and charge interactions with Cu_(2)Se,leading to the introduction of built-in electric field on heterojunction to improve electron mobility.Significantly,the theoretical calculations discovered that the underlying cause for the relatively superior rate capability in SIBs to that in PIBs is the agile Na~+diffusion with low energy barrier and moderate adsorption energy.These findings offer theoretical support for in-depth understanding of the performance differences of Cu-based materials in different ion storage systems.
文摘针对红外图像纹理弱及多目标遮挡导致跟踪精度低的问题,构建了基于改进YOLOv7模型和多目标跟踪算法DeepSort的融合红外目标跟踪模型MSB-YOLOv7-DeepSort。采用SE(squeeze and excitation)通道注意力机制和双向特征金字塔网络提高红外目标的特征提取质量;利用轻量化网络MobileNetV3替换YOLOv7骨干网络,提升融合模型的推理速度。实验结果表明,MSB-YOLOv7-DeepSort模型在跟踪准确度、跟踪精确度、正确目标跟踪比例和帧率等方面均具有较好的性能。
