3D Triangular Mesh Parameterization with Semantic Features Based on Competitive Learning Methods

We proposed a texture mapping technique that comprises mesh partitioning, mesh parameterization and packing, texture transferring, and texture correction and optimization for generating a high-quality texture map of a three-dimensional (3D) model for applications in e-commerce presentations. The main problems in texture mapping are that the texture resolution is generally worse than in the original images and considerable photo inconsistency exists at the transition of different image sources. To improve the texture resolution, we employed an oriented boundary box method for placing mesh islands on the parametric (UV) map. We also provided a texture size that can keep the texture resolution of the 3D textured model similar to that of the object images. To improve the photo inconsistency problem, we employed a method to detect and overcome the missing color that might exist on a texture map. We also proposed a blending process to minimize the transition error caused by different image sources. Thus, a high-quality 3D textured model can be obtained by applying this series of processes for presentations in e-commerce.

With 3D data rapidly emerging as an important form of multimedia information, 3D human mesh recovery technology has also advanced accordingly. However, current methods mainly focus on handling humans wearing tight clothing and perform poorly when estimating body shapes and poses under diverse clothing, especially loose garments. To this end, we make two key insights: (1) tailoring clothing to fit the human body can mitigate the adverse impact of clothing on 3D human mesh recovery, and (2) utilizing human visual information from large foundational models can enhance the generalization ability of the estimation. Based on these insights, we propose ClothHMR, to accurately recover 3D meshes of humans in diverse clothing. ClothHMR primarily consists of two modules: clothing tailoring (CT) and FHVM-based mesh recovering (MR). The CT module employs body semantic estimation and body edge prediction to tailor the clothing, ensuring it fits the body silhouette. The MR module optimizes the initial parameters of the 3D human mesh by continuously aligning the intermediate representations of the 3D mesh with those inferred from the foundational human visual model (FHVM). ClothHMR can accurately recover 3D meshes of humans wearing diverse clothing, precisely estimating their body shapes and poses. Experimental results demonstrate that ClothHMR significantly outperforms existing state-of-the-art methods across benchmark datasets and in-the-wild images. Additionally, a web application for online fashion and shopping powered by ClothHMR is developed, illustrating that ClothHMR can effectively serve real-world usage scenarios. The code and model for ClothHMR are available at: https://github.com/starVisionTeam/ClothHMR.

We present a parameterized representation of virtual organs for surgery simulation purpose. Random 3D input mesh are parameterized and resampled into a regular 2D parameterized model. With this parameterized representation, a high resolution 3D organ mesh can be reconstructed and deformed interactively with a simple and fast free-form deformation method. The amount of deformation and force feed-back can be calculated rapidly. Therefore, haptic rendering can be achieved. In addition, the parameterized mesh can be used to handle collision detection and the contact between multi-objects in an efficient way. With the parameterized mesh, realistic visual and haptic rendering can be provided for interactive surgery simulation.KeywordsCollision DetectionHaptic DeviceVirtual OrganLaparoscopic Heller MyotomyDeformation MethodThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Mesh parameterization is one of the fundamental operations in computer graphics (CG) and computeraided design (CAD). In this paper, we propose a novel local/global parameterization approach, ARAP++, for singleand multi-boundary triangular meshes. It is an extension of the as-rigid-as-possible (ARAP) approach, which stitches together 1-ring patches instead of individual triangles. To optimize the spring energy, we introduce a linear iterative scheme which employs convex combination weights and a fitting Jacobian matrix corresponding to a prescribed family of transformations. Our algorithm is simple, efficient, and robust. The geometric properties (angle and area) of the original model can also be preserved by appropriately prescribing the singular values of the fitting matrix. To reduce the area and stretch distortions for high-curvature models, a stretch operator is introduced. Numerical results demonstrate that ARAP++ outperforms several state-of-the-art methods in terms of controlling the distortions of angle, area, and stretch. Furthermore, it achieves a better visualization performance for several applications, such as texture mapping and surface remeshing.

Cryogenic electron tomography (cryo-ET) visualizes the 3D spatial distribution of macromolecules at nanometer resolution inside native cells. However, automated identification of macromolecules inside cellular tomograms is challenged by noise and reconstruction artifacts, as well as the presence of many molecular species in the crowded volumes. Here, we present DeepFinder, a computational procedure that uses artificial neural networks to simultaneously localize multiple classes of macromolecules. Once trained, the inference stage of DeepFinder is faster than template matching and performs better than other competitive deep learning methods at identifying macromolecules of various sizes in both synthetic and experimental datasets. On cellular cryo-ET data, DeepFinder localized membrane-bound and cytosolic ribosomes (roughly 3.2 MDa), ribulose 1,5-bisphosphate carboxylase-oxygenase (roughly 560 kDa soluble complex) and photosystem II (roughly 550 kDa membrane complex) with an accuracy comparable to expert-supervised ground truth annotations. DeepFinder is therefore a promising algorithm for the semiautomated analysis of a wide range of molecular targets in cellular tomograms.

This article explores the application of artificial intelligence (AI) technology in Go level examinations and its impact on Go education and evaluation systems. With the advancement of science and technology, especially the rapid development of AI technology, Go, as an ancient strategic game, has experienced revolutionary changes in its competitive scenes and learning methods. After AlphaGo defeated world Go champion Lee Sedol in 2016, the application of AI in the field of Go received widespread attention. AI has not only demonstrated capabilities beyond humans in high-level Go games, but has also played an increasingly important role in Go teaching and training. The global spread of the COVID-19 epidemic has forced Go level examinations to find new ways to adapt. The Go online intelligent test came into being, providing a new, efficient and objective assessment method. Although the introduction of online intelligent examinations has brought convenience, it has also raised issues in organizational structure adjustment, revenue distribution, and examination fairness. Challenges faced by the Go online intelligence examination in terms of fairness, including lack of face-to-face supervision, possible cheating, etc., pose new challenges to the fairness and authority of the Go level examination. This article compares the differences between intelligent Go exams and traditional exams, including assessment format, exam content and passing rate. The online intelligent examination has added test questions and raised the requirements for candidates’ computing power. It also shows a higher passing rate than traditional methods. However, the authenticity and fairness of online intelligent examinations have been challenged, including the adaptability of AI game difficulty and the lack of examination supervision, leading to the emergence of cheating. The positive impacts of implementing online smart exams include improving the convenience and efficiency of exams, simplifying event organization, and promoting the development of Go players’ skills. But at the same time, this new form has had an impact on the original Go organizational structure and benefit distribution, causing concerns and adjustments from local associations. In order to solve these problems, this article puts forward a series of suggestions, including setting up a supervision mechanism for intelligent examinations, optimizing the assessment system, and adjusting the revenue distribution ratio to ensure the long-term sustainable development of online intelligent examinations. In summary, the use of artificial intelligence in Go level examinations not only reflects the persistence of the interaction between technology and traditional cultural activities, but also poses ongoing challenges to the fairness and validity of education and assessment methods. By deeply understanding these impacts and taking corresponding countermeasures, we can find a balance between innovation and tradition, and provide support for the development of Go education and evaluation systems.

Multi-label learning aims at assigning a set of appropriate labels to multi-label samples. Although it has been successfully applied in various domains in recent years, most multi-label learning methods require sufficient labeled training samples, because of the large number of possible label sets. Co-training, as an important branch of semi-supervised learning, can leverage unlabeled samples, along with scarce labeled ones, and can potentially help with the large labeled data requirement. However, it is a difficult challenge to combine multi-label learning with co-training. Two distinct issues are associated with the challenge: (i) how to solve the widely-witnessed class-imbalance problem in multi-label learning; and (ii) how to select samples with confidence, and communicate their predicted labels among classifiers for model refinement. To address these issues, we introduce an approach called Multi-Label Co-Training (MLCT). MLCT leverages information concerning the co-occurrence of pairwise labels to address the class-imbalance challenge; it introduces a predictive reliability measure to select samples, and applies label-wise filtering to confidently communicate labels of selected samples among co-training classifiers. MLCT performs favorably against related competitive multi-label learning methods on benchmark datasets and it is also robust to the input parameters.

Inspired by the human learning principle that learning easier concepts first and then gradually paying more attention to harder ones, curriculum learning uses the nonuniform sampling of mini-batches according to the order of examples’ difficulty. Just as a teacher adjusts the curriculum according to the learning progress of each student, a proper curriculum should be adapted to the current state of the model. Therefore, in contrast to recent works using a fixed curriculum, we devise a new curriculum learning method, Adaptive Curriculum Learning (Adaptive CL), adapting the difficulty of examples to the current state of the model. Specifically, we make use of the loss of the current model to adjust the difficulty score while retaining previous useful learned knowledge by KL divergence. Moreover, under a non-linear model and binary classification, we theoretically prove that the expected convergence rate of curriculum learning monotonically decreases with respect to the loss of a point regarding the optimal hypothesis, and monotonically increases with respect to the loss of a point regarding the current hypothesis. The analyses indicate that Adaptive CL could improve the convergence properties during the early stages of learning. Extensive experimental results demonstrate the superiority of the proposed approach over existing competitive curriculum learning methods.

Many data mining tasks require the unsupervised partitioning of a data set into clusters. However, in many case we do not really know any prior knowledge about the clusters, for example, the density or the shape. This paper addresses two major issues associated with conventional competitive learning, namely, sensitivity to initialization and difficulty in determining the number of clusters. Many methods exist for such clustering, but most of then have assumed hyper-ellipsoidal clusters. Many heuristically proposed competitive learning methods and its variants, are somewhat ad hoc without any theoretical support. Under above considerations, we propose an algorithm named as Entropy guided Splitting Competitive Learning (ESCL) in the information theorem framework. Simulations show that minimization of partition entropy can be used to guide the competitive learning process, so to estimate the number and structure of probable data generators.

The analysis and forecasting of seasonal streamflow series are of utmost importance in the operation planning of water resource systems. One of the greatest difficulties in forecasting these series is the seasonal nature of streamflow series, due to wet and dry periods of the year. Moreover, real-world data are noisy and may contain contradictions and imperfections. Tolerance to imprecision and uncertainty is also required in order to achieve tractability and robustness. Fuzzy set-based data analysis models are especially suitable for these purposes. This suggests the application of neurofuzzy network models to seasonal streamflow forecasting. In this paper, a class of neurofuzzy network is applied to the problem of seasonal streamflow forecasting. This model is based on a constructive, competitive learning method where neuron groups compete when the network receives a new input, so that it learns the essential parameters to model a fuzzy system, which are the fuzzy rules and membership functions. A database of the average monthly inflows from a Brazilian hydroelectric plant was used. The performance of the model was compared with conventional approaches and the results show that the proposed model has a better performance than other methodologies which consider one-step-ahead forecasting and multi-step-ahead forecasting.

China has the largest number of hearing-impaired people in the worldover millions of Chinese suffered from hearing-impaired. However, most television programs don't offer Chinese subtitles or the sign language, leaving the hearing disabled people unable to obtain the information and enjoy recreational activities as ordinary people do. Therefore, the sign language on television is very important. Comparing to the ordinary computer, the calculation and processing power of the digital TV terminals is nowhere near a computer. Small storage space and low performance restrict the ability of displaying 3D graphics on the television. Therefore, this article uses a 3D graphics mesh simplification method which based on edge collapse, to simplified 3D graphics, and apply it to the sign language on the digital television, to achieve the goal of displaying the 3D sign language graphics smoothly and intuitively on the digital TV terminals.

Estimating 3D poses and shapes in the form of meshes from monocular RGB images is challenging. Obviously, it is more difficult than estimating 3D poses only in the form of skeletons or heatmaps. When interacting persons are involved, the 3D mesh reconstruction becomes more challenging due to the ambiguity introduced by person-to-person occlusions. To tackle the challenges, we propose a coarse-to-fine pipeline that benefits from 1) inverse kinematics from the occlusion-robust 3D skeleton estimation and 2) Transformer-based relation-aware refinement techniques. In our pipeline, we first obtain occlusion-robust 3D skeletons for multiple persons from an RGB image. Then, we apply inverse kinematics to convert the estimated skeletons to deformable 3D mesh parameters. Finally, we apply the Transformer-based mesh refinement that refines the obtained mesh parameters considering intra- and inter-person relations of 3D meshes. Via extensive experiments, we demonstrate the effectiveness of our method, outperforming state-of-the-arts on 3DPW, MuPoTS and AGORA datasets.KeywordsMulti-person3D mesh reconstructionTransformer

The dynamic competitive learning method

Even in the POST-PC era, the need for 3D graphics has steadily increased. In the design of the 3D graphics accelerator for mobile devices, the physical limitation such as power, size and input device must be considered. Moreover, the 3D graphics accelerator should reside in a single chip (SoC) with other blocks such as CPU, audio, video decoder and peripheral blocks. This work describes the design and implementation of mobile 3D graphics accelerator platform. We propose a mobile 3D graphics accelerator platform that has good performance and small size, and supports standard 3D graphics API (application programming interface). Proposed 3D graphics accelerator platform supports hardware Goraud shading, texture mapping, back & front face culling, Z buffering and alpha blending. This platform was verified on a FPGA test board with standard 3D API. It is suitable for cost effective mobile 3D platform such as smart phone, PDA (personal digital assistant), and other mobile devices.

We propose a fast and simple method for generating a low-stretch mesh parameterization. Given a triangle mesh, we start from the floater shape preserving parameterization and then improve the parameterization gradually. At each improvement step, we optimize the parameterization generated at the previous step by minimizing a weighted quadratic energy where the weights are chosen in order to minimize the parameterization stretch. This optimization procedure does not generate triangle flips if the boundary of the parameter domain is a convex polygon. Moreover already the first optimization step produces a high-quality mesh parameterization. We compare our parameterization procedure with several state-of-the-art mesh parameterization methods and demonstrate its speed and high efficiency in parameterizing large and geometrically complex models.