Background External knowledge representations play an essential role in knowledge-based visual question and answering to better understand complex scenarios in the open world.Recent entity-relationship embedding appro...Background External knowledge representations play an essential role in knowledge-based visual question and answering to better understand complex scenarios in the open world.Recent entity-relationship embedding approaches are deficient in representing some complex relations,resulting in a lack of topic-related knowledge and redundancy in topic-irrelevant information.Methods To this end,we propose MKEAH:Multimodal Knowledge Extraction and Accumulation on Hyperplanes.To ensure that the lengths of the feature vectors projected onto the hyperplane compare equally and to filter out sufficient topic-irrelevant information,two losses are proposed to learn the triplet representations from the complementary views:range loss and orthogonal loss.To interpret the capability of extracting topic-related knowledge,we present the Topic Similarity(TS)between topic and entity-relations.Results Experimental results demonstrate the effectiveness of hyperplane embedding for knowledge representation in knowledge-based visual question answering.Our model outperformed state-of-the-art methods by 2.12%and 3.24%on two challenging knowledge-request datasets:OK-VQA and KRVQA,respectively.Conclusions The obvious advantages of our model in TS show that using hyperplane embedding to represent multimodal knowledge can improve its ability to extract topic-related knowledge.展开更多
针对等矩柱状投影(equirectangular projection,ERP)全景视频多功能视频编码(versatile video coding,VVC)帧内编码复杂度过高的问题,提出一种编码单元(coding unit,CU)快速划分算法。根据ERP采样特性,将编码帧分为不同纬度区域。基于...针对等矩柱状投影(equirectangular projection,ERP)全景视频多功能视频编码(versatile video coding,VVC)帧内编码复杂度过高的问题,提出一种编码单元(coding unit,CU)快速划分算法。根据ERP采样特性,将编码帧分为不同纬度区域。基于不同纬度区域CU四叉树深度的分布特性和相邻CU的相关性,对当前CU的划分模式进行提前终止决策;利用梯度差异评估当前CU纹理特性,跳过冗余的水平或垂直划分模式。针对纹理模糊CU,通过纬度采样权重加权的二次比较,判断是否跳过垂直划分模式;利用二维哈尔小波变换系数评估当前CU子块间的差异,判断是否跳过三叉树划分模式。实验结果表明,在全帧内模式下,与VVC官方测试平台相比,所提算法能节省43.85%的编码时间,码率仅增加0.85%,视频质量没有明显下降。展开更多
The N-1 criterion is a critical factor for ensuring the reliable and resilient operation of electric power distribution networks.However,the increasing complexity of distribution networks and the associated growth in ...The N-1 criterion is a critical factor for ensuring the reliable and resilient operation of electric power distribution networks.However,the increasing complexity of distribution networks and the associated growth in data size have created a significant challenge for distribution network planners.To address this issue,we propose a fast N-1 verification procedure for urban distribution networks that combines CIM file data analysis with MILP-based mathematical modeling.Our proposed method leverages the principles of CIM file analysis for distribution network N-1 analysis.We develop a mathematical model of distribution networks based on CIM data and transfer it into MILP.We also take into account the characteristics of medium voltage distribution networks after a line failure and select the feeder section at the exit of each substation with a high load rate to improve the efficiency of N-1 analysis.We validate our approach through a series of case studies and demonstrate its scalability and superiority over traditional N-1 analysis and heuristic optimization algorithms.By enabling online N-1 analysis,our approach significantly improves the work efficiency of distribution network planners.In summary,our proposed method provides a valuable tool for distribution network planners to enhance the accuracy and efficiency of their N-1 analyses.By leveraging the advantages of CIM file data analysis and MILP-based mathematical modeling,our approach contributes to the development of more resilient and reliable electric power distribution networks.展开更多
基金Supported by National Nature Science Foudation of China(61976160,61906137,61976158,62076184,62076182)Shanghai Science and Technology Plan Project(21DZ1204800)。
文摘Background External knowledge representations play an essential role in knowledge-based visual question and answering to better understand complex scenarios in the open world.Recent entity-relationship embedding approaches are deficient in representing some complex relations,resulting in a lack of topic-related knowledge and redundancy in topic-irrelevant information.Methods To this end,we propose MKEAH:Multimodal Knowledge Extraction and Accumulation on Hyperplanes.To ensure that the lengths of the feature vectors projected onto the hyperplane compare equally and to filter out sufficient topic-irrelevant information,two losses are proposed to learn the triplet representations from the complementary views:range loss and orthogonal loss.To interpret the capability of extracting topic-related knowledge,we present the Topic Similarity(TS)between topic and entity-relations.Results Experimental results demonstrate the effectiveness of hyperplane embedding for knowledge representation in knowledge-based visual question answering.Our model outperformed state-of-the-art methods by 2.12%and 3.24%on two challenging knowledge-request datasets:OK-VQA and KRVQA,respectively.Conclusions The obvious advantages of our model in TS show that using hyperplane embedding to represent multimodal knowledge can improve its ability to extract topic-related knowledge.
文摘针对等矩柱状投影(equirectangular projection,ERP)全景视频多功能视频编码(versatile video coding,VVC)帧内编码复杂度过高的问题,提出一种编码单元(coding unit,CU)快速划分算法。根据ERP采样特性,将编码帧分为不同纬度区域。基于不同纬度区域CU四叉树深度的分布特性和相邻CU的相关性,对当前CU的划分模式进行提前终止决策;利用梯度差异评估当前CU纹理特性,跳过冗余的水平或垂直划分模式。针对纹理模糊CU,通过纬度采样权重加权的二次比较,判断是否跳过垂直划分模式;利用二维哈尔小波变换系数评估当前CU子块间的差异,判断是否跳过三叉树划分模式。实验结果表明,在全帧内模式下,与VVC官方测试平台相比,所提算法能节省43.85%的编码时间,码率仅增加0.85%,视频质量没有明显下降。
基金supported by the National Natural Science Foundation of China(52207105)。
文摘The N-1 criterion is a critical factor for ensuring the reliable and resilient operation of electric power distribution networks.However,the increasing complexity of distribution networks and the associated growth in data size have created a significant challenge for distribution network planners.To address this issue,we propose a fast N-1 verification procedure for urban distribution networks that combines CIM file data analysis with MILP-based mathematical modeling.Our proposed method leverages the principles of CIM file analysis for distribution network N-1 analysis.We develop a mathematical model of distribution networks based on CIM data and transfer it into MILP.We also take into account the characteristics of medium voltage distribution networks after a line failure and select the feeder section at the exit of each substation with a high load rate to improve the efficiency of N-1 analysis.We validate our approach through a series of case studies and demonstrate its scalability and superiority over traditional N-1 analysis and heuristic optimization algorithms.By enabling online N-1 analysis,our approach significantly improves the work efficiency of distribution network planners.In summary,our proposed method provides a valuable tool for distribution network planners to enhance the accuracy and efficiency of their N-1 analyses.By leveraging the advantages of CIM file data analysis and MILP-based mathematical modeling,our approach contributes to the development of more resilient and reliable electric power distribution networks.
