Explore the vast range of titles available.

In this article, we introduce Tsinghua Global Minimum （TGMin） as a new program for the global minimum searching of geometric structures of gas-phase or surface-supported atomic clusters, and the constrained basin-hopping （BH） algorithm implemented in this program. To improve the efficiency of the BH algorithm, several types of constraints are introduced to reduce the vast search space, including constraints on the random displacement step size, displacement of low-coordination atoms, and geometrical structure adjustment after displacement. The ultrafast shape-recognition （USR） algorithm and its variants are implemented to identify duplicate structures during the global minimum search. In addition to the Metropolis acceptance criterion, we also implemented a morphology-based constraint that confines the global minimum search to a specific type of morphology, such as planar or non-planar structures, which offers a strict divide-and-conquer strategy for the BH algorithm. These improvements are implemented in the TGMin program, which was developed over the past decade and has been used in a number of publications. We tested our TGMin program on global minimum structural searches for a number of metal and main-group clusters including C60, Au20 and B20 clusters. Over the past five years, the TGMin program has been used to determine the global minimum structures of a series of boron atomic clusters （such as [B26]^-, [B28]^-, [B30]^-, [B35]^-, [B36]^-, [B39]^-, [B40]^-, [MnB16]^-, [COB18]^-, [RhB18]^-, and [TaB20]^-）, metal-containing clusters Lin （n = 3-20）, Aug（CO）8^＋ and [Cr6O19]^2-. and the oxide-supported metal catalyst Au7/γ-Al2O3, as well as other isolated and surface-supported atomic clusters. In this article we present the major features of TGMin program and show that it is highly efficient at searching for global-minimum structures of atomic clusters in the gas phase and on various surface supports.

In many industrial applications, it is possible to approximate the shape of mechanical parts with geometric primitives such as spheres, boxes, and cylinders. This information can be used to plan robotic grasping and manipulation procedures. The work presented in this paper investigated the use of the state-of-the-art PointNet deep neural network for primitive shape recognition in 3D scans of real-life objects. To obviate the need of collecting a large set of training models, it was decided to train PointNet using examples generated from artificial geometric models. The motivation of the study was the achievement of fully automated disassembly operations in remanufacturing applications. PointNet was chosen due to its suitability to process 3D models, and ability to recognise objects irrespective of their poses. The use of simpler shallow neural network procedures was also evaluated. Twenty-eight point cloud scenes of everyday objects selected from the popular Yale-CMU-Berkeley benchmark model set were used in the experiments. Experimental evidence showed that PointNet is able to generalise the knowledge gained on artificial shapes, to recognise shapes in ordinary objects with reasonable accuracy. However, the experiments showed some limitations in this ability of generalisation, in terms of average accuracy (78% circa) and consistency of the learning procedure. Using a feature extraction procedure, a multi-layer-perceptron architecture was able to achieve nearly 83% classification accuracy. A practical solution was proposed to improve PointNet generalisation capabilities: by training the neural network using an error-corrupted scene, its accuracy could be raised to nearly 86%, and the consistency of the learning results was visibly improved.

3D point cloud recognition is fundamental and popular in vision perceptual systems such as autonomous driving, robotics, and virtual reality. Due to the sparse distribution and irregularity of point clouds, previous 3D point networks perform convolution on nearby points, ignoring the long-range dependence on the global structure. To solve this problem, we propose a Viewport Group Point Cloud Network for 3D Shape Recognition (VGPCNet) in which features are grouped according to viewports instead of local neighbor points to model the long-range global context. First, we propose to use viewport as proxy to capture both local and global features from an outside view of the object. The related points are grouped by visibility attribute effectively and efficiently which can not only capture the inside local geometry details but also obtain the global structure from the outside viewport. Second, we use a graph-based feature consolidation module to enhance the viewport features by modeling interactions between different viewports. Finally, to aggregate a global representation from multiple viewport features, we propose a novel attention-based feature aggregation module. We evaluate our VGPCNet on three widely used benchmarks including ModelNet40/10, ScanObjectNN, and ShapeCore55 for shape classification and retrieval tasks. Extensive experiments have demonstrated the effectiveness and superior performance (94.1% on ModelNet40) of our method over state-of-the-art methods.

The recognition and classification of building shapes are the prerequisites and foundation for building simplification, matching, and change detection, which have always been important research problems in the field of cartographic generalization. Due to the ambiguity and uncertainty of building shape outlines, it is difficult to describe them using unified rules, which has always limited the quality and automation level of building shape recognition. In response to the above issues, by introducing object detection technology in computer vision, this article proposes a building shape recognition and classification method based on the YOLO object detection model. Firstly, for different types of buildings, four levels of building training data samples are constructed, and YOLOv5, YOLOv8, YOLOv9, and YOLOv9 integrating attention modules are selected for training. The trained models are used to test the shape judgment of buildings in the dataset and verify the learning effectiveness of the models. The experimental results show that the YOLO model can accurately classify and locate the shape of buildings, and its recognition and detection effect have the ability to simulate advanced human visual cognition, which provides a new solution for the fuzzy shape recognition of buildings with complex outlines and local deformation.

3D shape recognition with few examples is crucial for applications involving 3D scenes, but typical methods based on surface and view suffer the failure to describe the interior and exterior features uniformly. Thus, we propose 3D orthogonal integral transform (OIT). OIT is composed of three individual integrals over a group of three orthogonal planes rotating to cover all orientations by which the volumetric shape is bisected in integrals. OIT offers the following advantages: (1) It describes a 3D shape structurally from interior to exterior uniformly, which brings about discriminative shape characteristics; and (2) the shape descriptor built on OIT is invariant with respect to translation, scaling and rotation. Furthermore, a fine-grained 3D model dataset (FGModele40) is built on ModelNet40. Experiments show that OIT can provide both discriminative and robust descriptors for 3D shape recognition with few examples. Our proposed OIT outperforms typical state-of-the-art benchmarks evaluated by the protein shape retrieval contest; additionally, it also surpasses other typical deep learning models with respect to the task of 3D shape recognition with few examples on FGModele40.

The mode shapes of many beam-type structures, such as aircraft wings and wind turbine blades, involve a high degree of coupling between flap-wise and edge-wise bending deformations. In the case of wind turbine blades, the principal bending deformations (flap-wise deformation and edge-wise deformation) are typically easily recognized by visual observation. However, this visual approach is sometimes challenging for high-order mode shapes that have complex mode deformation. More importantly, visual observation cannot quantify the contribution of different deformation components for each mode. This work proposes a novel mode shape recognition algorithm, called Mode Shape Recognition Matrix (MSRM), for application to complicated spatial beam-type structures not only to identify the deformation components of the complex beam mode shapes, but more importantly, to quantify their respective relative contribution. In the application case studied for the MSRM method, a three-dimensional wind turbine blade is mapped into three-dimensional Chebyshev polynomial space. The blade mode shape is correlated to each polynomial shape function to find the contribution of each polynomial shape function for the mode shape. To validate the mode shape recognition performance, MSRM is applied on both numerical mode shapes from a fidelity blade finite element model and experimental mode shapes from a high spatial resolution 3D SLDV modal test. Both numerical and experimental studies demonstrate that MSRM can successfully recognize the quantitative contribution of flap-wise deformation and edge-wise deformation for each wind turbine blade mode shape.

Adopting many viewpoints and mining the relationship between them, 3D shape recognition inferring the object’s category from 2D rendered images has proven effective. However, using a limited number of general representative viewpoints to form a reasonable expression of the object is a task with both practical and theoretical significance. This paper proposes a multi-view CNN architecture with independent viewpoint feature extraction and the unity of importance weights, which can dramatically decrease the number of viewpoints by learning the representative ones. First, the view-based and independent view features are extracted by a deep neural network. Second, the network automatically learns relativity between these viewpoints and outputs the importance weights of views. Finally, view features are aggregated to predict the category of objects. Through iterative learning of these critical weights in instances, global representative viewpoints are selected. We assess our method on two challenging datasets, ModelNet and ShapeNet. Rigorous experiments show that our strategy is competitive with the latest method using only six viewpoints and RGB information as input. Meanwhile, our approach also achieves state-of-the-art performance by using 20 viewpoints as input. Specifically, the proposed approach achieves 99.34% and 97.49% accuracy on the ModelNet10 and ModelNet40, and 80.0% mAP on ShapeNet.

The Multi-view Convolution Neural Network (MVCNN) has achieved considerable success in 3D shape recognition. However, 3D shape recognition using view-images from random viewpoints has not been yet exploited in depth. In addition, 3D shape recognition using a small number of view-images remains difficult. To tackle these challenges, we developed a novel Multi-view Convolution Neural Network, “Latent-MVCNN” (LMVCNN), that recognizes 3D shapes using multiple view-images from pre-defined or random viewpoints. The LMVCNN consists of three types of sub Convolution Neural Networks. For each view-image, the first type of CNN outputs multiple category probability distributions and the second type of CNN outputs a latent vector to help the first type of CNN choose the decent distribution. The third type of CNN outputs the transition probabilities from the category probability distributions of one view to the category probability distributions of another view, which further helps the LMVCNN to find the decent category probability distributions for each pair of view-images. The three CNNs cooperate with each other to the obtain satisfactory classification scores. Our experimental results show that the LMVCNN achieves competitive performance in 3D shape recognition on ModelNet10 and ModelNet40 for both the pre-defined and the random viewpoints and exhibits promising performance when the number of view-images is quite small.

As an essential field of multimedia and computer vision, 3D shape recognition has attracted much research attention in recent years. Multiview‐based approaches have demonstrated their superiority in generating effective 3D shape representations. Typical methods usually extract the multiview global features and aggregate them together to generate 3D shape descriptors. However, there exist two disadvantages: First, the mainstream methods ignore the comprehensive exploration of local information in each view. Second, many approaches roughly aggregate multiview features by adding or concatenating them together. The information loss for some discriminative characteristics limits the representation effectiveness. To address these problems, a novel architecture named region‐based joint attention network (RJAN) was proposed. Specifically, the authors first design a hierarchical local information exploration module for view descriptor extraction. The region‐to‐region and channel‐to‐channel relationships from different granularities can be comprehensively explored and utilised to provide more discriminative characteristics for view feature learning. Subsequently, a novel relation‐aware view aggregation module is designed to aggregate the multiview features for shape descriptor generation, considering the view‐to‐view relationships. Extensive experiments were conducted on three public databases: ModelNet40, ModelNet10, and ShapeNetCore55. RJAN achieves state‐of‐the‐art performance in the tasks of 3D shape classification and 3D shape retrieval, which demonstrates the effectiveness of RJAN. The code has been released on https://github.com/slurrpp/RJAN.

This study aims to realize a precision grasp of unknown-shaped objects. Precision grasping requires a detailed understanding of the surface shapes such as concavity and convexity. If an accurate shape model is not given in advance, it must be addressed by sensing. We have proposed a method for recognizing detailed object shapes using proximity sensors equipped on each fingertip of a multi-fingered robot hand. Direct sensing of the object’s surface from the fingertips enables both avoidance of unintended collision during the approach process and recognition of surface profiles for use in planning and executing stable grasping. This paper introduces local surface curvature estimation to improve the accuracy of local surface recognition. We propose practical and accurate models to estimate local curvature based on various characteristic tests on the proximity sensor and to estimate the distance to the nearest point. In actual experiments, it was shown that it was possible to estimate the position of the nearest point with a mean error of less than 2 mm and to predict grasping stability in reasonable real-time for the object shape.