Dynamic Scene Graph Generation of Point Clouds with Structural Representation Learning

Scene graphs of point clouds help to understand object-level relationships in the 3D space.Most graph generation methods work on 2D structured data,which cannot be used for the 3D unstructured point cloud data.Existing point-cloud-based methods generate the scene graph with an additional graph structure that needs labor-intensive manual annotation.To address these problems,we explore a method to convert the point clouds into structured data and generate graphs without given structures.Specifically,we cluster points with similar augmented features into groups and establish their relationships,resulting in an initial structural representation of the point cloud.Besides,we propose a Dynamic Graph Generation Network(DGGN)to judge the semantic labels of targets of different granularity.It dynamically splits and merges point groups,resulting in a scene graph with high precision.Experiments show that our methods outperform other baseline methods.They output reliable graphs describing the object-level relationships without additional manual labeled data.

Phase Representation and Property Determination of Raw Materials of Solid Lubricant

The composition, microstructure, mechanical and frictional properties of PTFE and its fillers were represented and analyzed by XRD, SEM, DSC, XPS and large-scale polarizing microscope. The results show that PTFE has a flocculent structure with high melt temperature and decomposition temperature, big contact angle and crystallinity, and low surface hardness, compression strength, friction coefficient, wearing capacity and surface energy. Cooling rate influenced the friction coefficient and wear resistance. Graphite and molybdenum disulfide have a flake structure, and molybdenum disulfide has a big contact angle and low surface energy. Copper powder has a globular structure and its chief component is Cu-Pb alloy, and there is a loose layer on the surface. Carbon fiber has a rod structure and there are C=O and C-O-C polar groups on the skeleton surface. The decreasing order of water absorption capacity is graphite, carbon fiber, molybdenum disulfide, PTFE and copper powder.

Heuristic Backtrack Algorithm for Structural Match and Its Applications

The concept WALKING on structures is proposed, and the partial ordering between a structure and a query structure (substructure) is also created by means of WALKING. Based upon the above concepts, authors create the Heuristic-Backtracking Algorithm (HBA) of structural match with high performance. In the last part of the paper, the applications of HBA in molecular graphics, synthetic planning, spectrum simulation , the representation and recognition of general structures are discussed.

Free form deformation and symmetry constraint‐based multimodal brain image registration using generative adversarial nets

Multi‐modal brain image registration has been widely applied to functional localisation,neurosurgery and computational anatomy.The existing registration methods based on the dense deformation fields involve too many parameters,which is not conducive to the exploration of correct spatial correspondence between the float and reference images.Meanwhile,the unidirectional registration may involve the deformation folding,which will result in the change of topology during registration.To address these issues,this work has presented an unsupervised image registration method using the free form deformation(FFD)and the symmetry constraint‐based generative adversarial networks(FSGAN).The FSGAN utilises the principle component analysis network‐based structural representations of the reference and float images as the inputs and uses the generator to learn the FFD model parameters,thereby producing two deformation fields.Meanwhile,the FSGAN uses two discriminators to decide whether the bilateral registration have been realised simultaneously.Besides,the symmetry constraint is utilised to construct the loss function,thereby avoiding the deformation folding.Experiments on BrainWeb,high grade gliomas,IXI and LPBA40 show that compared with state‐of‐the‐art methods,the FSGAN provides superior performance in terms of visual comparisons and such quantitative indexes as dice value,target registration error and computational efficiency.

STATE:Learning structure and texture representations for novel view synthesis

Novel viewpoint image synthesis is very challenging,especially from sparse views,due to large changes in viewpoint and occlusion.Existing image-based methods fail to generate reasonable results for invisible regions,while geometry-based methods have difficulties in synthesizing detailed textures.In this paper,we propose STATE,an end-to-end deep neural network,for sparse view synthesis by learning structure and texture representations.Structure is encoded as a hybrid feature field to predict reasonable structures for invisible regions while maintaining original structures for visible regions,and texture is encoded as a deformed feature map to preserve detailed textures.We propose a hierarchical fusion scheme with intra-branch and inter-branch aggregation,in which spatio-view attention allows multi-view fusion at the feature level to adaptively select important information by regressing pixel-wise or voxel-wise confidence maps.By decoding the aggregated features,STATE is able to generate realistic images with reasonable structures and detailed textures.Experimental results demonstrate that our method achieves qualitatively and quantitatively better results than state-of-the-art methods.Our method also enables texture and structure editing applications benefiting from implicit disentanglement of structure and texture.Our code is available at http://cic.tju.edu.cn/faculty/likun/projects/STATE.

An Improved Graphic Representation for Structured Program Design

In this paper,an improved graphic representation for Structured Program Design——N-S-Z (Nassi-Shneiderman-Zhou Diagram)is proposed.It not only preserves the advantages of the conventional graphic and non-graphic representations,but also adds some new features which will enhance the representa- tive power of the original diagram.

Topology optimization of support structure of telescope skin based on bit-matrix representation NSGA-II

Non-dominated sorting genetic algorithm II（NSGA-II）with multiple constraints handling is employed for multi-objective optimization of the topological structure of telescope skin,in which a bit-matrix is used as the representation of a chromosome,and genetic algorithm（GA）operators are introduced based on the matrix.Objectives including mass,in-plane performance,and out-of-plane load-bearing ability of the individuals are obtained by fnite element analysis（FEA）using ANSYS,and the matrix-based optimization algorithm is realized in MATLAB by handling multiple constraints such as structural connectivity and in-plane strain requirements.Feasible confgurations of the support structure are achieved.The results confrm that the matrix-based NSGA-II with multiple constraints handling provides an effective method for two-dimensional multi-objective topology optimization.