A new algorithm for the solution of under constraint graph in sketch drawing is put forward. The directed process of constraint graph is completed by picking concealed constraints of adjacent entities in sketch of few...A new algorithm for the solution of under constraint graph in sketch drawing is put forward. The directed process of constraint graph is completed by picking concealed constraints of adjacent entities in sketch of few or no dimensions. In this paper, the priority of concealed constraint is given by the different constraint types and constructing orders, and some more priority concealed constraints are forced into obvious ones by the need number of constraint for every node in constraint solution process.展开更多
A promising approach on the search for geometric constraint of directionalgraph in off-line parametric design has been put forward. The method for representation of thedirectional graph is given firstly, then the cons...A promising approach on the search for geometric constraint of directionalgraph in off-line parametric design has been put forward. The method for representation of thedirectional graph is given firstly, then the constraint model for every geometry element is putforward. By the search of geometric constraint, it sets up the topology constraint relation on thedifferent element objects, which are not related in geometry data base, and achieves the identifyfor the directional graph. The algorithm of recognizing geometric constraint is described in detail,in the meantime, the application example in off-line parametric design on dimension-driven isgiven.展开更多
This paper presents one novel spatial geometric constraints histogram descriptors （SGCHD） based on curvature mesh graph for automatic three-dimensional （3D） pollen particles recognition. In order to reduce high di...This paper presents one novel spatial geometric constraints histogram descriptors （SGCHD） based on curvature mesh graph for automatic three-dimensional （3D） pollen particles recognition. In order to reduce high dimensionality and noise disturbance arising from the abnormal record approach under microscopy, the separated surface curvature voxels are ex- tracted as primitive features to represent the original 3D pollen particles, which can also greatly reduce the computation time for later feature extraction process. Due to the good invariance to pollen rotation and scaling transformation, the spatial geometric constraints vectors are calculated to describe the spatial position correlations of the curvature voxels on the 3D curvature mesh graph. For exact similarity evaluation purpose, the bidirectional histogram algorithm is applied to the spatial geometric constraints vectors to obtain the statistical histogram descriptors with fixed dimensionality, which is invariant to the number and the starting position of the curvature voxels. Our experimental results compared with the traditional methods validate the argument that the presented descriptors are invariant to different pollen particles geometric transformations （such as posing change and spatial rotation）, and high recognition precision and speed can be obtained simultaneously.展开更多
在知识图谱补全领域,实体和关系之间丰富的多语义信息对于提升补全任务的准确性具有重要意义。然而,现有模型往往难以充分捕捉和整合这些多语义特征,导致补全效果受限。为此,提出了一种基于多语义提取的知识图谱补全模型(Knowledge Grap...在知识图谱补全领域,实体和关系之间丰富的多语义信息对于提升补全任务的准确性具有重要意义。然而,现有模型往往难以充分捕捉和整合这些多语义特征,导致补全效果受限。为此,提出了一种基于多语义提取的知识图谱补全模型(Knowledge Graph Completion Model Based on Multi-Semantic Extraction,MSE)。首先,设计了多语义聚合编码器,对实体和关系嵌入进行维度切分,整合了邻居实体和关系的多语义信息。其次,设计了基于多尺度卷积的解码器,使用不同大小的卷积核提取实体自身的深层语义特征。最后,设计了融合独立性约束的损失函数,引入基于Pearson相关系数的正则化项,以提升模型多语义表达能力。实验结果表明,在FB15k-237和WN18RR数据集上,相较于其他基线的最优模型,MSE模型的MRR(Mean Reciprocal Rank)值分别提升1.7%和2.3%,验证了其在知识图谱补全任务上的有效性。展开更多
文摘A new algorithm for the solution of under constraint graph in sketch drawing is put forward. The directed process of constraint graph is completed by picking concealed constraints of adjacent entities in sketch of few or no dimensions. In this paper, the priority of concealed constraint is given by the different constraint types and constructing orders, and some more priority concealed constraints are forced into obvious ones by the need number of constraint for every node in constraint solution process.
基金This project is supported by National 863/C1MS Plan(No.863-511-507) Provincial Natural Science Foundation of Anhui, China (No.01042209).
文摘A promising approach on the search for geometric constraint of directionalgraph in off-line parametric design has been put forward. The method for representation of thedirectional graph is given firstly, then the constraint model for every geometry element is putforward. By the search of geometric constraint, it sets up the topology constraint relation on thedifferent element objects, which are not related in geometry data base, and achieves the identifyfor the directional graph. The algorithm of recognizing geometric constraint is described in detail,in the meantime, the application example in off-line parametric design on dimension-driven isgiven.
基金supported by the National Natural Science Foundation of China(Grant No.61375030)the Natural Science Foundation of Jiangsu Province,China(Grant No.BK20090149)the Natural Science Foundation of Higher Education Institutions of Jiangsu Province,China(Grant No.08KJD520019)
文摘This paper presents one novel spatial geometric constraints histogram descriptors （SGCHD） based on curvature mesh graph for automatic three-dimensional （3D） pollen particles recognition. In order to reduce high dimensionality and noise disturbance arising from the abnormal record approach under microscopy, the separated surface curvature voxels are ex- tracted as primitive features to represent the original 3D pollen particles, which can also greatly reduce the computation time for later feature extraction process. Due to the good invariance to pollen rotation and scaling transformation, the spatial geometric constraints vectors are calculated to describe the spatial position correlations of the curvature voxels on the 3D curvature mesh graph. For exact similarity evaluation purpose, the bidirectional histogram algorithm is applied to the spatial geometric constraints vectors to obtain the statistical histogram descriptors with fixed dimensionality, which is invariant to the number and the starting position of the curvature voxels. Our experimental results compared with the traditional methods validate the argument that the presented descriptors are invariant to different pollen particles geometric transformations （such as posing change and spatial rotation）, and high recognition precision and speed can be obtained simultaneously.
文摘在知识图谱补全领域,实体和关系之间丰富的多语义信息对于提升补全任务的准确性具有重要意义。然而,现有模型往往难以充分捕捉和整合这些多语义特征,导致补全效果受限。为此,提出了一种基于多语义提取的知识图谱补全模型(Knowledge Graph Completion Model Based on Multi-Semantic Extraction,MSE)。首先,设计了多语义聚合编码器,对实体和关系嵌入进行维度切分,整合了邻居实体和关系的多语义信息。其次,设计了基于多尺度卷积的解码器,使用不同大小的卷积核提取实体自身的深层语义特征。最后,设计了融合独立性约束的损失函数,引入基于Pearson相关系数的正则化项,以提升模型多语义表达能力。实验结果表明,在FB15k-237和WN18RR数据集上,相较于其他基线的最优模型,MSE模型的MRR(Mean Reciprocal Rank)值分别提升1.7%和2.3%,验证了其在知识图谱补全任务上的有效性。
