Search

Abstract Utilizing Generative Adversarial Networks (GANs) to generate 3D representations of the Terracotta Warriors offers a novel approach for the preservation and restoration of cultural heritage. Through GAN technology, we can produce complete 3D models of the Terracotta Warriors’ faces, aiding in the repair of damaged or partially destroyed figures. This paper proposes a distillation model, DIGAN, for generating 3D Terracotta Warrior faces. By extracting knowledge from StyleGAN2, we train an innovative 3D generative network. G2D, the primary component of the generative network, produces detailed and realistic 2D images. The 3D generator modularly decomposes the generation process, covering texture, shape, lighting, and pose, ultimately rendering 2D images of the Terracotta Warriors’ faces. The model enhances the learning of 3D shapes through symmetry constraints and multi-view data, resulting in high-quality 2D images that closely resemble real faces. Experimental results demonstrate that our method outperforms existing GAN-based generation methods.

Abstract The bulk density of grains from a celestial body is a fundamental property related to its composition and structure, contributing to the understanding of its evolutionary history. In this study, we provide the bulk density of 637 grains returned from the C-type near-Earth asteroid 162173 Ryugu. This is the largest number of grains to date for the curation activity, corresponding to 38 wt.% of the total returned samples (approximately 5.4 g). Although several densities of the Ryugu grains were reported, the volume estimation of some samples showed uncertainties. Therefore, we applied a new volume estimation model calibrated by X-ray micro-computed tomography (XCT) to the Ryugu grains to more accurately estimate their bulk density. The obtained average bulk density of 637 Ryugu grains was 1.79 ± 0.31 g/cm3 (1σ variation) for weights of 0.5‒100 mg (sub-mm ‒to 10 mm) irrespective of their 3D shapes characterized by three axial length ratios, considered to be a representative of the returned samples. The bulk density distributions of the grains in Chambers A and C were statistically distinguishable, with mean values of 1.81 ± 0.30 and 1.76 ± 0.33 g/cm3 (1σ variations), respectively. Despite the small difference, bulk density may have differed by sampling site. The obtained average bulk density value of A + C samples was almost the same as that of 16 Ryugu grains estimated based on CT scanned data, and was consistent with the densities of CI chondrites (1.57–1.91 g/cm3). The axial ratios of the grains in Chambers A and C were similar and those of the 724 returned samples and the flying particles ejected during the sampling operations were also similar, suggesting that relatively small Ryugu materials (mm‒cm in size) are similar in shape. The minor difference between the Rygu grains and flying particles could be attributed to events such as scraping during sampling operations and transportation. Graphical Abstract

A variety of saliency detection techniques have been proposed during the last two decades to determine important regions on the surface of 3D shapes in form of triangular meshes. However, most fail in predicting the regions where human eyes naturally fixate when observing and exploring an object. Taking inspiration from biological studies that enumerate a list of object characteristics revealed in human visual processing and the influence of semantic properties in the emergence of neural responses in human brain, in this work, we propose a deep convolutional neural network architecture using gradient-based class activation mapping to detect saliencies on the surface of 3D objects when classifying them based on their different properties. We further argue that using Pearson Correlation Coefficient is not sufficient for the evaluation of saliency values and therefore propose a novel evaluation technique to determine how reliable is the detection performed by saliency detectors to predict eye fixations. More specifically, this evaluation metric measures the distance between the most salient region detected and the respective location of human eye fixation. Evaluating the results based on visual comparison, as well as using the proposed evaluation technique, demonstrates that our model is successful in predicting the locations where human eye fixates. Results are compared with five state-of-the-art saliency detectors, and our experiments suggest that in average the location of the highest saliency detected by our approach is closer to the location of human eye fixation by about 22.55% to 77.76% in comparison with five state-of-the art method" on a public dataset. [ABSTRACT FROM AUTHOR]

Multi-view learning methods have achieved remarkable results in 3D shape recognition. However, most of them focus on the visual feature extraction and feature aggregation, while viewpoints (spatial positions of virtual cameras) for generating multiple views are often ignored. In this paper, we deeply explore the correlation between viewpoints and shape descriptor, and propose a novel viewpoint-guided prototype learning network (VGP-Net). We introduce a prototype representation for each class, including viewpoint prototype and feature prototype. The viewpoint prototype is the average weight of each viewpoint learned from a small support set via Score Unit, and stored in a weight dictionary. Our VGP model self-adaptively learns the view-wise weights by dynamically assembling with the viewpoint prototypes in weight dictionary and performing element-wise operation via view pooling layer. Under the guidance of viewpoint prototype, important visual features are enhanced, while those negligible features are eliminated. These refined features are effectively fused to generate compact shape descriptor. All the shape descriptors are clustered in feature embedding space, and the cluster center represents the feature prototype of each class. The classification thus can be performed by searching the nearest distance to feature prototypes. To boost the learning process, we further present a multi-stream regularization mechanism in both feature space and viewpoint space. Extensive experiments demonstrate that our VGP-Net is efficient, and the learned deep features have stronger discrimination ability. Therefore, it can achieve better performance compared to state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The bulk density of grains from a celestial body is a fundamental property related to its composition and structure, contributing to the understanding of its evolutionary history. In this study, we provide the bulk density of 637 grains returned from the C-type near-Earth asteroid 162173 Ryugu. This is the largest number of grains to date for the curation activity, corresponding to 38 wt.% of the total returned samples (approximately 5.4 g). Although several densities of the Ryugu grains were reported, the volume estimation of some samples showed uncertainties. Therefore, we applied a new volume estimation model calibrated by X-ray micro-computed tomography (XCT) to the Ryugu grains to more accurately estimate their bulk density. The obtained average bulk density of 637 Ryugu grains was 1.79 ± 0.31 g/cm3 (1σ variation) for weights of 0.5‒100 mg (sub-mm ‒to 10 mm) irrespective of their 3D shapes characterized by three axial length ratios, considered to be a representative of the returned samples. The bulk density distributions of the grains in Chambers A and C were statistically distinguishable, with mean values of 1.81 ± 0.30 and 1.76 ± 0.33 g/cm3 (1σ variations), respectively. Despite the small difference, bulk density may have differed by sampling site. The obtained average bulk density value of A + C samples was almost the same as that of 16 Ryugu grains estimated based on CT scanned data, and was consistent with the densities of CI chondrites (1.57–1.91 g/cm3). The axial ratios of the grains in Chambers A and C were similar and those of the 724 returned samples and the flying particles ejected during the sampling operations were also similar, suggesting that relatively small Ryugu materials (mm‒cm in size) are similar in shape. The minor difference between the Rygu grains and flying particles could be attributed to events such as scraping during sampling operations and transportation. [ABSTRACT FROM AUTHOR]

The global pandemic has changed the mode of delivering lessons. Adult learners have missed and continue to miss the hands-on learning activity, which is a form of experiential learning. When it comes to teaching three-dimensional, geometric shapes (3D), learners often benefit from handling and manipulating concrete shapes. This study examined the use of an AR mobile application, as a learning tool for teaching geometric shapes in virtual mathematics classrooms. Data was collected from thirty learners at Functional Skills and General Certificate of Secondary Education (GCSE) levels in a Further Education college in Birmingham, UK. Learners were divided into Control and Experimental groups. The Control group used traditional learning materials, while in the Experimental group, an AR application was utilised during lessons. Qualitative data was gathered using mixed methods of one-to-one interviews and questionnaires to evaluate the learners' experience. The data was triangulated with observation and note-taking by the teacher. In this study, the content analysis method has been used to analyse the data. Results showed an improved activity and interactivity within the session, when compared to the traditional, passive, teaching and learning strategies used in the learning process. From the results of this study, we can deduce that AR-based tools are a beneficial resource in teaching mathematics and improving learner experience in virtual classrooms. Further studies are required to evaluate quantitative improvements in student performance, as well as to implement AR as a regular teaching strategy. [ABSTRACT FROM AUTHOR]

Copyright of Industria Textila is the property of Institutul National de Cercetare-Dezvoltare pentru Textile si Pielarie and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Hydrogel shape memory and actuating functionalities are heavily pursued and have found great potential in various application fields. However, their combination for more flexible and complicated morphing behaviors is still challenging. Herein, it is reported that by controlling the light‐initiated polymerization of active hydrogel layers on shape memory hydrogel substrates, advanced morphing behaviors based on programmable hydrogel shapes and actuating trajectories are realized. The formation and photo‐reduction‐induced dissociation of Fe3+‐carboxylate coordination endow the hydrogel substrates with the shape memory functionality. The photo‐reduced Fe2+ ions can diffuse from the substrates into the monomer solutions to initiate the polymerization of the thermally responsive active layers, whose actuating temperatures and amplitudes can be facially tuned by controlling their thicknesses and compositions. One potential application, a shape‐programmable 3D hook that can lift an object with a specific shape, is also unveiled. The demonstrated strategy is extendable to other hydrogel systems to realize more versatile and complicated actuating behaviors. [ABSTRACT FROM AUTHOR]

Today's consumers use a range of cues to identify product origin, including brand names, logos, colours and shapes. The range of registrable marks has therefore expanded, but this brings a risk that features which others have a legitimate competitive need to use will fall under the exclusive control of a single undertaking. Registration may also be used to extend the finite duration of other IP rights that the owner has already enjoyed. Consequently, trade mark law contains functionality limitations on registration designed to protect these competitive concerns. This piece considers how well those limitations are working. The CJEU has also seemed to apply stricter distinctiveness rules to shape marks based on the assumption that consumers are not used to seeing shapes as origin indicators. Some have assumed this means that it is almost impossible to register shape marks—this research examines whether this is really so. How functionality and distinctiveness work in practice is examined through an empirical analysis over a 5-year period of all shape mark applications to the European Union Intellectual Property Office. This piece considers which types of marks are being registered, which are being refused and why. It reveals that distinctiveness, rather than functionality, is having the biggest impact on shape mark registration, and in fact a larger number of shape marks than expected are registered: often because of the addition of non-3D matter. However, are a significant number of marks comprised just of product shapes. It concludes with a discussion of competitive and policy challenges identified by this research. [ABSTRACT FROM AUTHOR]

This article describes a tetrahedral kite activity that was implemented with grade 9 students (age 14-15). We detail how the three-part lesson provided opportunities to integrate mathematics and art concepts, with potential to also weave in science and engineering ideas. The first part primed students to consider tetrahedral kites, their cultural and historical significance, and the materials needed for constructing the kite. The second part had students create a prototype using nets of tissue paper decorated with mark making techniques. The third part had students create a tetrahedron kite containing cultural and geographical mark making techniques on the tissue paper sides before flying the kites at a community event. We conclude the article with recommendations to help other teachers integrate mathematics and visual arts topics through tetrahedral kites. [ABSTRACT FROM AUTHOR]

Cryogenic electron microscopy (cryo-EM) has become widely used for the past few years in structural biology, to collect single images of macromolecules "frozen in time". As this technique facilitates the identification of multiple conformational states adopted by the same molecule, a direct product of it is a set of 3D volumes, also called EM maps. To gain more insights on the possible mechanisms that govern transitions between different states, and hence the mode of action of a molecule, we recently introduced a bioinformatic tool that interpolates and generates morphing trajectories joining two given EM maps. This tool is based on recent advances made in optimal transport, that allow efficient evaluation of Wasserstein barycenters of 3D shapes. As the overall performance of the method depends on various key parameters, including the sensitivity of the regularization parameter, we performed various numerical experiments to demonstrate how MorphOT can be applied in different contexts and settings. Finally, we discuss current limitations and further potential connections between other optimal transport theories and the conformational heterogeneity problem inherent with cryo-EM data.

Most existing 3D geometry copy detection research focused on 3D watermarking, which first embeds "watermarks" and then detects the added watermarks. However, this kind of methods is non-straightforward and may be less robust to attacks such as cropping and noise. In this paper, we focus on a fundamental and practical research problem: judging whether a point cloud is plagiarized or copied to another point cloud in the presence of several manipulations (e.g., similarity transformation, smoothing). We propose a novel method to address this critical problem. Our key idea is first to align the two point clouds and then calculate their similarity distance. We design three different measures to compute the similarity. We also introduce two strategies to speed up our method. Comprehensive experiments and comparisons demonstrate the effectiveness and robustness of our method in estimating the similarity of two given 3D point clouds. [ABSTRACT FROM AUTHOR]

We present 3DPointCaps++ for learning robust, flexible and generalizable 3D object representations without requiring heavy annotation efforts or supervision. Unlike conventional 3D generative models, our algorithm aims for building a structured latent space where certain factors of shape variations, such as object parts, can be disentangled into independent sub-spaces. Our novel decoder then acts on these individual latent sub-spaces (i.e. capsules) using deconvolution operators to reconstruct 3D points in a self-supervised manner. We further introduce a cluster loss ensuring that the points reconstructed by a single capsule remain local and do not spread across the object uncontrollably. These contributions allow our network to tackle the challenging tasks of part segmentation, part interpolation/replacement as well as correspondence estimation across rigid / non-rigid shape, and across / within category. Our extensive evaluations on ShapeNet objects and human scans demonstrate that our network can learn generic representations that are robust and useful in many applications. [ABSTRACT FROM AUTHOR]

3D shape clustering is developing into an important research subject with the wide applications of 3D shapes in computer vision and multimedia fields. Since 3D shapes generally take on various modalities, how to comprehensively exploit the multi-modal properties to boost clustering performance has become a key issue for the 3D shape clustering task. Taking into account the advantages of multiple views and point clouds, this paper proposes the first multi-modal 3D shape clustering method, named the dual contrastive learning network (DCL-Net), to discover the clustering partitions of unlabeled 3D shapes. First, by simultaneously performing cross-view contrastive learning within multi-view modality and cross-modal contrastive learning between the point cloud and multi-view modalities in the representation space, a representation-level dual contrastive learning module is developed, which aims to capture discriminative 3D shape features for clustering. Meanwhile, an assignment-level dual contrastive learning module is designed by further ensuring the consistency of clustering assignments within the multi-view modality, as well as between the point cloud and multi-view modalities, thus obtaining more compact clustering partitions. Experiments on two commonly used 3D shape benchmarks demonstrate the effectiveness of the proposed DCL-Net. [ABSTRACT FROM AUTHOR]

Existing methods for extracting 3D curve skeletons mostly suffer from the difficulty of finding the center points of 3D shapes and tedious manual adjustments of the thresholds for pruning spurious branches due to the influence of shape boundary perturbations. In this article, we present a method based on medial surfaces of 3D shapes for the convenient and fast extraction of high-quality curve skeletons. Our main contribution is a simple and stable centeredness measure. It is based on simulating fire propagation via the scheme of inside-out evolution from the interior to the boundary, differentiating it from existing methods that use the scheme of outside-in evolution from the boundary to the interior. Thus, our measure is much more localized, and it can be implemented with a high degree of parallelism. In addition, we propose measures to effectively suppress the influence of details to obtain a stable measurement, and employ minimum set covers to optimize the center points to generate compact skeletons, which enables spurious branches to be effectively excluded without the tedious work of manually adjusting thresholds. Our experiments show the superiority of our method over existing methods, including its convenient generation of clean, compact and centered curve skeletons while running much faster than state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The developed technologies for creating the three-dimensional shape for the formation of feminine personal protective armor proved to be applicable in the case of high-performance protective triaxial fabric, made of polyester, Vectran, and Kevlar. The suggested method combines cyclic loading and deep forming under heat. Cup shape geometry was analyzed to get the relations between the cup volume and the cup diameter for different cup bra sizing systems. This work analyzed the factors of molding triaxial fabric in creating a three-dimensional shape to get a bra fitting essential for the comfort of the protective armor and to provide useful information about the treatment of different fabrics molding. A smoothly formed cup surface has a natural configuration of the breast shape and also provides unlimited designs. The suggested technology of fabric forming will architect fabric from 2D to construct the required 3D shapes. In this work, a setup was designed for heating the fabric accomplished with the application of variable cyclic loading. The procedure of forming bra cup shapes with the deep drawing that permits 3D fabric architects for the bra shapes of varied sizes was given. [ABSTRACT FROM AUTHOR]

Summary: This paper proposes a new method for generating 3D objects based on generative adversarial networks (GANs). Recently, GANs have been used in 3D object generation, but it is still very challenging to generate high‐quality 3D objects because of the complex data distribution over 3D objects. In this paper, we propose a system based on GAN that makes the generated objects more realistic. We use multiple generators and discriminators to enhance the ability of the model for learning complex distributions. Such a stacked structure can be considered as a coarse‐to‐fine or low‐to‐high–resolution mechanism. We employ the spectral normalization technology to control the Lipschitz constant of the discriminators by literally constraining the spectral norm of each layer to get a more stable training process. In this way, the proposed model can generate realistic and high‐quality 3D objects. Moreover, our system can also recover incomplete 3D objects into complete 3D objects. Experiments demonstrate that our model performs better in the quality of the generated objects than the baselines. [ABSTRACT FROM AUTHOR]

3D shape clustering is developing into an important research subject with the wide applications of 3D shapes in computer vision and multimedia fields. Since 3D shapes generally take on various modalities, how to comprehensively exploit the multi-modal properties to boost clustering performance has become a key issue for the 3D shape clustering task. Taking into account the advantages of multiple views and point clouds, this paper proposes the first multi-modal 3D shape clustering method, named the dual contrastive learning network (DCL-Net), to discover the clustering partitions of unlabeled 3D shapes. First, by simultaneously performing cross-view contrastive learning within multi-view modality and cross-modal contrastive learning between the point cloud and multi-view modalities in the representation space, a representation-level dual contrastive learning module is developed, which aims to capture discriminative 3D shape features for clustering. Meanwhile, an assignment-level dual contrastive learning module is designed by further ensuring the consistency of clustering assignments within the multi-view modality, as well as between the point cloud and multi-view modalities, thus obtaining more compact clustering partitions. Experiments on two commonly used 3D shape benchmarks demonstrate the effectiveness of the proposed DCL-Net.

Measuring the intrinsic deformability of arbitrary small‐scale subdivision of a shape is an interesting meanwhile valuable research topic. Such measurement can be directly utilised as a reliable criteria to partition shape into small components and then assist in shape modelling and description. Compared to global modelling, through constructing subdivision‐based complex shape description, the accuracy and flexibility of shape representation can be significantly improved. In this study, the authors propose a line segment advection (LSA)‐based vertex‐level three‐dimensional shape deformability measuring method. It can highlight the deformability characteristics of each shape part in any scale and size. The measurement is realised mainly based on the advection of line segments connecting neighbouring shape mesh vertices. For 3D shapes, since the line segment of triangular mesh facet directly reflects the minimal neighbourhood relationships and mesh microstructure, its advection can capture the finest details of shape deformability. Then, after transferring that information into neighbouring vertices, a vertex‐level shape deformability measurement can be acquired. Besides, to demonstrate the value of the proposed measuring method to shape partitioning and piecewise shape modelling, a straightforward shape partitioning method is introduced as well. Extensive experiments on three publicly available databases are conducted to verify the effectiveness of proposed methods.

3D mesh segmentation has become an essential step in many applications in 3D shape analysis. In this paper, a new segmentation method is proposed based on a learning approach using the artificial neural networks classifier and the spectral clustering for segmentation. Firstly, a training step is done using the artificial neural network trained on existing segmentation, taken from the ground truth segmentation (done by humane operators) available in the benchmark proposed by Chen et al. to extract the candidate boundaries of a given 3D-model based on a set of geometric criteria. Then, we use this resulted knowledge to construct a new connectivity of the mesh and use the spectral clustering method to segment the 3D mesh into significant parts. Our approach was evaluated using different evaluation metrics. The experiments confirm that the proposed method yields significantly good results and outperforms some of the competitive segmentation methods in the literature.

Given the important role played by students' spatial reasoning skills, in this paper we analyse how students use these skills to solve problems involving 2D representations of 3D geometrical shapes. Using data from in total 1357 grades 4 to 9 students, we examine how they visualise shapes in the given diagrams and make use of properties of shapes to reason. We found that using either spatial visualisation or property-based spatial analytic reasoning is not enough for the problems that required more than one step of reasoning, but also that these two skills have to be harmonised by domain-specific knowledge in order to overcome the perceptual appearance (or "look") of the given diagram. We argue that more opportunities might be given to both primary and secondary school students in which they can exercise not only their spatial reasoning skills but also consolidate and use their existing domain-specific knowledge of geometry for productive reasoning in geometry. [ABSTRACT FROM AUTHOR]

Considering the fact that points of interest on 3D shapes can be discriminated from a geometric perspective, it is reasonable to map the geometric signature of a point $p$p to a probability value encoding to what degree $p$p is a point of interest, especially for a specific class of 3D shapes. Based on the observation, we propose a three-phase algorithm for learning and predicting points of interest on 3D shapes by using multiple feature descriptors. Our algorithm requires two separate deep neural networks (stacked auto-encoders) to accomplish the task. During the first phase, we predict the membership of the given 3D shape according to a set of geometric descriptors using a deep neural network. After that, we train the other deep neural network to predict a probability distribution defined on the surface representing the possibility of a point being a point of interest. Finally, we use a manifold clustering technique to extract a set of points of interest as the output. Experimental results show superior detection performance of the proposed method over the previous state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

Background: Chorea-acanthocytosis (ChAc) is a rare hereditary neurodegenerative disease with deformed red blood cells (RBCs), so-called acanthocytes, as a typical marker of the disease. Erythrocyte sedimentation rate (ESR) was recently proposed as a diagnostic biomarker. To date, there is no treatment option for affected patients, but promising therapy candidates, such as dasatinib, a Lyn-kinase inhibitor, have been identified. Methods: RBCs of two ChAc patients during and after dasatinib treatment were characterized by the ESR, clinical hematology parameters and the 3D shape classification in stasis based on an artificial neural network. Furthermore, mathematical modeling was performed to understand the contribution of cell morphology and cell rigidity to the ESR. Microfluidic measurements were used to compare the RBC rigidity between ChAc patients and healthy controls. Results: The mechano-morphological characterization of RBCs from two ChAc patients in an off-label treatment with dasatinib revealed differences in the ESR and the acanthocyte count during and after the treatment period, which could not directly be related to each other. Clinical hematology parameters were in the normal range. Mathematical modeling indicated that RBC rigidity is more important for delayed ESR than cell shape. Microfluidic experiments confirmed a higher rigidity in the normocytes of ChAc patients compared to healthy controls. Conclusions: The results increase our understanding of the role of acanthocytes and their associated properties in the ESR, but the data are too sparse to answer the question of whether the ESR is a suitable biomarker for treatment success, whereas a correlation between hematological and neuronal phenotype is still subject to verification.

Ultimately, any information to be processed by computers must be represented as a plain sequence of numbers. This poses a major challenge when representing complex 3D shapes. Marcel Campen's research explores how to efficiently identify a good quad layout (a partition of a surface into simple four-sided pieces). This article shows how novel algorithmic ideas can be combined to form both automatic and interactive quad-layout generation pipelines. [ABSTRACT FROM PUBLISHER]

Flexible definition and automatic extraction of generic features on 3D shapes is important for feature-centric geometric analysis, however, existing techniques fall short in measuring and locating semantic features from users' psychological standpoints. This paper makes an initial attempt to propose a learning based generic modeling approach for user-central definition and automatic extraction of features on arbitrary shapes. Instead of purely resorting to certain local geometric extremes to simply formulate feature metrics, it enables the users to arbitrarily specify application-specific features on training shapes, so that similar features can be automatically extracted from other same-category shapes with isometric or near-isometric deformations. Our key originality is built upon an observation: the geodesic distance from one point to desired feature point on testing shape should be similar to the points on training shapes. To this end, we propose a novel regression model to bridge the massive random-sampled local properties and the desired feature points via incorporating their corresponding geodesic distances into the powerful random forest framework. On that basis, an effective voting strategy is proposed to estimate the locations of the user-specified features on new shapes. Our extensive experiments and comprehensive evaluations have demonstrated many attractive advantages of our method, including being fully-automatic, robust to noise and partial holes, invariant under isometric and near-isometric deformation, and also scale-invariant, which can well facilitate to many downstream geometry-processing applications such as semantic mesh segmentation, mesh skeleton extraction, etc. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel unsupervised algorithm for automatically segmenting a single 3D shape or co-segmenting a family of 3D shapes using deep learning. The algorithm consists of three stages. In the first stage, we pre-decompose each 3D shape of interest into primitive patches to generate over-segmentation and compute various signatures as low-level shape features. In the second stage, high-level features are learned, in an unsupervised style, from the low-level ones based on deep learning. Finally, either segmentation or co-segmentation results can be quickly reported by patch clustering in the high-level feature space. The experimental results on the Princeton Segmentation Benchmark and the Shape COSEG Dataset exhibit superior segmentation performance of the proposed method over the previous state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

Making use of semantic information to achieve the high-level analysis and understanding is a hot issue currently. To address the problem of automatic recognition in the presence of significant geometric and topological variations, this paper proposed a 3D shape function recognition method by adopting the contextual relationship of shape parts. Firstly, it decomposed 3D shapes into the shape part sets with independent semantics and employed the technique of approximate convexity analysis.Then, it computed the contextual relationship of shape parts and on this basis, and adopted support vector machines to achieve the task of automatic recognition between shape parts. Experimental results show that the proposed method achieves higher the matching accuracy values and lower classification error rates, compared to the existing methods. [ABSTRACT FROM AUTHOR]