Collapse Modes Research Articles

AI-driven automatic generation and rendering of game characters

Abstract. This paper provides a comprehensive review of AI-driven techniques for the automatic generation and rendering of game characters, with a particular focus on Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs). GANs, using adversarial training approaches, have revolutionized character development by producing incredibly realistic and aesthetically pleasing gaming characters. VAEs, despite frequently encountering issues like image blurriness, provide an alternate strategy that emphasizes diversity and originality in the generated content. Additionally, conditional models that enable more individualized and regulated character generation are explored, as well as hybrid models that combine the best features of both GANs and VAEs. The difficulties with mode collapse in GANs and the requirement for big datasets for both GANs and VAEs are also covered, along with some possible fixes like transfer learning and semi-supervised learning strategies. This analysis emphasizes the growing significance of AI-driven game character generation in the gaming industry by highlighting its current state, problems, and future directions.

An Empirical Study of WGAN and WGAN-GP for Enhanced Image Generation

This paper aims to advance the Wasserstein Generative Adversarial Networks (WGANs) and their enhancements, particularly focusing on the gradient penalty. Generative Adversarial Networks (GANs), introduced by Goodfellow et al. in 2014, have revolutionized the domain of image generation. To address the limitations of GANs, the WGAN was proposed. However, WGANs rely on weight clipping, which introduces its own set of issues such as slow convergence and potential gradient vanishing. The inefficiency and instability of WGANs have troubled its users. To solve these problems, WGAN with Gradient Penalty (WGAN-GP) was developed to address these challenges. It provides more stable gradients and reduces the risk of mode collapse by using a gradient penalty to enforce the necessary constraints. In this paper, the author implemented both WGAN and WGAN with Gradient Penalty (WGAN-GP) and evaluated them using the CIFAR-10 and MNIST datasets. The results show that WGAN-GP's outputs are more stable and efficient in the early rounds, confirming the effectiveness of the gradient penalty in training image datasets.

Crystal Structure Prediction Using Generative Adversarial Network with Data-Driven Latent Space Fusion Strategy.

Crystal structure prediction (CSP) is an important field of material design. Herein, we propose a novel generative adversarial network model, guided by a data-driven approach and incorporating the real physical structure of crystals, to address the complexity of high-dimensional data and improve prediction accuracy in materials science. The model, termed GAN-DDLSF, introduces a novel sampling method called data-driven latent space fusion (DDLSF), which aims to optimize the latent space of generative adversarial networks (GANs) by combining the statistical properties of real data with a standard Gaussian distribution, effectively mitigating the "mode collapse" problem prevalent in GANs. Our approach introduces a more refined generation mechanism specifically for binary crystal structures such as gallium nitride (GaN). By optimizing for the specific crystallographic features of GaN while maintaining structural rationality, we achieve higher precision and efficiency in predicting and designing structures for this particular material system. The model generates 9321 GaN binary crystal structures, with 16.59% reaching a stable state and 24.21% found to be metastable. These results can significantly enhance the accuracy of crystal structure predictions and provide valuable insights into the potential of the GAN-DDLSF approach for the discovery and design of binary, ternary, and multinary materials, offering new perspectives and methods for materials science research and applications.

DMFGAN: a multifeature data augmentation method for grape leaf disease identification.

The use of deep learning techniques to identify grape leaf diseases relies on large, high-quality datasets. However, a large number of images occupy more computing resources and are prone to pattern collapse during training. In this paper, a depth-separable multifeature generative adversarial network (DMFGAN) was proposed to enhance grape leaf disease data. First, a multifeature extraction block (MFEB) based on the four-channel feature fusion strategy is designed to improve the quality of the generated image and avoid the problem of poor feature learning ability of the adversarial generation network caused by the single-channel feature extraction method. Second, a depth-based D-discriminator is designed to improve the discriminator capability and reduce the number of model parameters. Third, SeLU activation function was substituted for DCGAN activation function to overcome the problem that DCGAN activation function was not enough to fit grape leaf disease image data. Finally, an MFLoss function with a gradient penalty term is proposed to reduce the mode collapse during the training of generative adversarial networks. By comparing the visual indicators and evaluation indicators of the images generated by different models, and using the recognition network to verify the enhanced grape disease data, the results show that the method is effective in enhancing grape leaf disease data. Under the same experimental conditions, DMFGAN generates higher quality and more diverse images with fewer parameters than other generative adversarial networks. The mode breakdown times of generative adversarial networks in training process are reduced, which is more effective in practical application.

MulCPred: Learning Multi-Modal Concepts for Explainable Pedestrian Action Prediction.

Pedestrian action prediction is crucial for many applications such as autonomous driving. However, state-of-the-art methods lack the explainability needed for trustworthy predictions. In this paper, a novel framework called MulCPred is proposed that explains its predictions based on multi-modal concepts represented by training samples. Previous concept-based methods have limitations, including the following: (1) they cannot be directly applied to multi-modal cases; (2) they lack the locality needed to attend to details in the inputs; (3) they are susceptible to mode collapse. These limitations are tackled accordingly through the following approaches: (1) a linear aggregator to integrate the activation results of the concepts into predictions, which associates concepts of different modalities and provides ante hoc explanations of the relevance between the concepts and the predictions; (2) a channel-wise recalibration module that attends to local spatiotemporal regions, which enables the concepts with locality; (3) a feature regularization loss that encourages the concepts to learn diverse patterns. MulCPred is evaluated on multiple datasets and tasks. Both qualitative and quantitative results demonstrate that MulCPred is promising in improving the explainability of pedestrian action prediction without obvious performance degradation. Moreover, by removing unrecognizable concepts, MulCPred shows improved cross-dataset prediction performance, suggesting its potential for further generalization.

Shear Strengthening of RC Beams Incorporating Post-Tensioned Bars and Engineered Cementitious Composite Reinforced with Palm Fronds

This paper investigates, experimentally and numerically, the shear strengthening of Normal Concrete (NC) beams using post-tensioning steel bars and Engineered Cementitious Composite (ECC) reinforced with chemically cured Palm Fronds (PFs). The benefits of strain-hardening ECC and the tensile strength of PFs cured with 6% wt Alkali NaOH solution beside post-tensioned bars have been employed herein. Seven full-scale Reinforced Concrete (RC) beams were fabricated and experimented with under three-point loading until failure. The test parameters include the strengthening technique, type, and configuration of the material used for strengthening. The strengthening process has been implemented through two techniques: Externally Bonded Reinforcement (EBR) and Near-Surface Mounted (NSM) Reinforcement. The strengthening materials have been configured and placed in horizontal, vertical, and inclined positions. The effectiveness of the strengthening methods has been evaluated by examining their cracking propagations, load-deflection responses, collapse modes, elastic stiffness, and absorbed energy. It was found that the proposed strengthening systems could significantly control the crack pattern and failure mode, and they could enhance the ultimate load amplitude up to 37% and 50% for NSM ECC with PFs and EBR post-tensioning steel bars, respectively. Nonlinear three-dimensional finite element models of the tested beams were developed and validated with the test data, where it was found that finite element models predict the structural performance of tested beams with a maximum error of only 2%.

Enhanced WGAN Model for Diagnosing Laryngeal Carcinoma

This study modifies the U-Net architecture for pixel-based segmentation to automatically classify lesions in laryngeal endoscopic images. The advanced U-Net incorporates five-level encoders and decoders, with an autoencoder layer to derive latent vectors representing the image characteristics. To enhance performance, a WGAN was implemented to address common issues such as mode collapse and gradient explosion found in traditional GANs. The dataset consisted of 8171 images labeled with polygons in seven colors. Evaluation metrics, including the F1 score and intersection over union, revealed that benign tumors were detected with lower accuracy compared to other lesions, while cancers achieved notably high accuracy. The model demonstrated an overall accuracy rate of 99%. This enhanced U-Net model shows strong potential in improving cancer detection, reducing diagnostic errors, and enhancing early diagnosis in medical applications.

Styled and characteristic Peking opera facial makeup synthesis with co-training and transfer conditional styleGAN2

Against the backdrop of the deep integration of culture and technology, research and practice in digitization of intangible cultural heritage has continued to deepen. However, due to the lack of data and training, it is still very difficult to apply artificial intelligence to the field of cultural heritage protection. This article integrates image generation technology into the digital protection of Peking opera facial makeup, using a self-built Peking opera facial makeup dataset. Based on the StyleGAN2 network, we propose a style generative cooperative training network Co-StyleGAN2, which integrates the adaptive data augmentation (ADA) to alleviate the problem of discriminator overfitting and introduces the idea of cooperative training to stabilize the training process. We design a Peking opera facial makeup image transform conditional generation network TC-StyleGAN2 which is transferred from unconditional generation network. The weights of the unconditional pre-training model are fixed, and an adaptive filtering modulation module is added to modulate the category parameters to complete the conversion from unconditional to conditional StyleGAN2 to deal with the training difficulty of conditional GANs on limited data, which suffer from severe mode collapse. The experimental results show that the proposed training strategy is better than the comparison algorithms, and the image generation quality and diversity have been improved.

A novel 8-connected Pixel Identity GAN with Neutrosophic (ECP-IGANN) for missing imputation

Missing pixel imputation presents a critical challenge in image processing and computer vision, particularly in applications such as image restoration and inpainting. The primary objective of this paper is to accurately estimate and reconstruct missing pixel values to restore complete visual information. This paper introduces a novel model called the Enhanced Connected Pixel Identity GAN with Neutrosophic (ECP-IGANN), which is designed to address two fundamental issues inherent in existing GAN architectures for missing pixel generation: (1) mode collapse, which leads to a lack of diversity in generated pixels, and (2) the preservation of pixel integrity within the reconstructed images. ECP-IGANN incorporates two key innovations to improve missing pixel imputation. First, an identity block is integrated into the generation process to facilitate the retention of existing pixel values and ensure consistency. Second, the model calculates the values of the 8-connected neighbouring pixels around each missing pixel, thereby enhancing the coherence and integrity of the imputed pixels. The efficacy of ECP-IGANN was rigorously evaluated through extensive experimentation across five diverse datasets: BigGAN-ImageNet, the 2024 Medical Imaging Challenge Dataset, the Autonomous Vehicles Dataset, the 2024 Satellite Imagery Dataset, and the Fashion and Apparel Dataset 2024. These experiments assessed the model’s performance in terms of diversity, pixel imputation accuracy, and mode collapse mitigation, with results demonstrating significant improvements in the Inception Score (IS) and Fréchet Inception Distance (FID). ECP-IGANN markedly enhanced image segmentation performance in the validation phase across all datasets. Key metrics, such as Dice Score, Accuracy, Precision, and Recall, were improved substantially for various segmentation models, including Spatial Attention U-Net, Dense U-Net, and Residual Attention U-Net. For example, in the 2024 Medical Imaging Challenge Dataset, the Residual Attention U-Net’s Dice Score increased from 0.84 to 0.90, while accuracy improved from 0.88 to 0.93 following the application of ECP-IGANN. Similar performance enhancements were observed with the other datasets, highlighting the model’s robust generalizability across diverse imaging domains.

Unpaired Image-to-Image Translation with Diffusion Adversarial Network

Unpaired image translation with feature-level constraints presents significant challenges, including unstable network training and low diversity in generated tasks. This limitation is typically attributed to the following situations: 1. The generated images are overly simplistic, which fails to stimulate the network’s capacity for generating diverse and imaginative outputs. 2. The images produced are distorted, a direct consequence of unstable training conditions. To address this limitation, the unpaired image-to-image translation with diffusion adversarial network (UNDAN) is proposed. Specifically, our model consists of two modules: (1) Feature fusion module: In this module, one-dimensional SVD features are transformed into two-dimensional SVD features using the convolutional two-dimensionalization method, enhancing the diversity of the images generated by the network. (2) Network convergence module: In this module, the generator transitions from the U-net model to a superior diffusion model. This shift leverages the stability of the diffusion model to mitigate the mode collapse issues commonly associated with adversarial network training. In summary, the CycleGAN framework is utilized to achieve unpaired image translation through the application of cycle-consistent loss. Finally, the proposed network was verified from both qualitative and quantitative aspects. The experiments show that the method proposed can generate more realistic converted images.

Pulsational mode dynamics in the presence of negions

The stability dynamics, of the gravito-electrostatically driven pulsational mode of gravitational collapse (PMGC) excitable in spherically symmetric dust molecular clouds (DMCs), is systematically theorized. It actively incorporates the interplay of dust viscosity, dust charge fluctuation dynamics, and negionic (negative ionic) effects. We take the DMC constitutive electrons, positive ions, and negions as the Boltzmann species (lighter). The negatively charged dust grains behave as viscous fluids (heavier). A generalized linear cubic dispersion equation, involving a new and unique set of multi-factorial dispersion coefficients, results from the applied spherical normal mode treatment in the DMC model. The diverse growth characteristics of the PMGC dynamics are investigated illustratively. It is found mainly that the cloud size, dust charge, equilibrium electronic number density, and equilibrium dust number density play stabilizing roles in the PMGC. However, the dust mass, equilibrium positive ionic number density, equilibrium negionic number density, and dust viscosity play destabilizing roles against self-gravity. The reliability of our study is ensured by the fact that the diversified stabilizing features explored here are consistent with the previous results. The applicability of our study in astrophysical gravity-driven structure formation processes moderated actively with the negionic effects is finally outlined.

VMP: Versatile Motion Priors for Robustly Tracking Motion on Physical Characters

AbstractRecent progress in physics‐based character control has made it possible to learn policies from unstructured motion data. However, it remains challenging to train a single control policy that works with diverse and unseen motions, and can be deployed to real‐world physical robots. In this paper, we propose a two‐stage technique that enables the control of a character with a full‐body kinematic motion reference, with a focus on imitation accuracy. In a first stage, we extract a latent space encoding by training a variational autoencoder, taking short windows of motion from unstructured data as input. We then use the embedding from the time‐varying latent code to train a conditional policy in a second stage, providing a mapping from kinematic input to dynamics‐aware output. By keeping the two stages separate, we benefit from self‐supervised methods to get better latent codes and explicit imitation rewards to avoid mode collapse. We demonstrate the efficiency and robustness of our method in simulation, with unseen user‐specified motions, and on a bipedal robot, where we bring dynamic motions to the real world.

Post impact flexural behavior investigation of hybrid foam-core sandwich composites at extreme Arctic temperature

This study explores the post-impact bending behavior and failure mechanisms in hybrid sandwich composites made of Carbon Fiber Reinforced Polymer (CFRP) and Glass Fiber Reinforced Polymer (GFRP). Flexural tests conducted at both ambient room temperature and low temperature Arctic conditions reveal a significant enhancement in flexural performance when GFRP layer is incorporated on the outer side of the hybrid composite. The investigation utilizes images from testing to elucidate damage modes, including fiber and matrix cracking in the composite facesheet, as well as core shearing and debonding in the Polyvinyl Chloride (PVC) foam core. Residual flexural properties are notably influenced by stacking sequence, facesheet compressive properties, pre-existing impact damage and temperature conditions. Analytical predictions, validated experimentally, highlight the effect of stacking sequence, low temperature, and impact energy on flexural collapse modes, with competing failure modes such as indentation and core shear. Collapse maps indicate that room temperature specimens predominantly collapse through indentation, while diverse collapse mechanisms emerge due to facesheet thickness, rigidity, and degraded tensile strength. The study aims to provide fundamental insights for future composite designs tailored for Arctic applications.

Instability and Evolution of Shocked Clouds Formed by Orthogonal Collisions between Magnetized Filamentary Molecular Clouds

Filamentary molecular clouds are recognized as primary sites for the formation of stars. Specifically, regions characterized by the overlapping point of multiple filaments, known as hub regions, are often associated with active star formation. However, the formation mechanism of this hub structure is not well understood. Therefore, to understand the formation mechanism and star formation in hub structures, as a first step, we investigate the orthogonal collisions between two filaments using three-dimensional ideal magnetohydrodynamical simulations. As a model of initial filaments, we use an infinitely long filament in magnetohydrostatic equilibrium under a global magnetic field running perpendicular to the filament axis. Two identical equilibrium filaments, sharing the same magnetic flux, are arranged with their long axes perpendicular to each other and given an initial velocity perpendicular to their long axes to replicate an orthogonal collision. We find three types of evolution after the shocked cloud is formed: collapse, stable, and expansion modes. The energy balance just after the filaments completely collide explains the future evolution of the shocked cloud. If the magnitude of gravitational energy is larger than the sum of the kinetic, thermal, and magnetic energies, the shocked cloud evolves in collapse mode. If the magnitude of gravitational energy is less than the sum of these energies, the cloud evolves in stable mode when the kinetic energy is relatively small and in expansion mode when the kinetic energy is sufficiently large.

Constructing three-dimension digital rock with porosity information constraint: A double-network-cycled style-based deep-learning approach

Understanding reservoir properties from geophysical responses necessitates the construction of accurate petrophysical templates based on adequate petrophysical data. However, engineering coring is often limited by complex subsurface conditions and high costs. Artificial intelligence (AI) techniques offer an efficient and economical way to synthesize digital samples. Nevertheless, traditional deep learning approaches may suffer from mode collapse, particularly when generating samples with complex structures from a limited number of training samples. To address this challenge, we propose novel Generative Adversarial Networks (GANs) to controllably generate 3D digital rock samples according to porosity distribution. By employing a style-transfer generator, multi-scale information of 3D digital rock is integrated into the generation process, effectively reducing the risk of mode collapse. Embedding the generator into a double-network-cycled framework further enhances the controllability of conditional information in the generated samples. Our analysis shows that the minimum error between the generated samples and the desired samples in terms of porosity is only 0.07%. A clear contrast is observed in morphological parameters, and differences in pore structure lead to significant variations in mechanical and hydraulic properties between original samples and synthetic samples with similar porosity. This indicates that the property contrast is likely caused by differences in pore structures rather than porosity. These findings will assist in future studies on the effect of pore structure on petrophysical properties and improve the utility of rock physics templates in geophysical inversion.

Multi-adversarial autoencoders: Stable, faster and self-adaptive representation learning

The variational autoencoder (VAE) and generative adversarial networks (GAN) are two prominent approaches to achieving a probabilistic generative model by way of an autoencoder and a two-player minimax game. While VAEs often suffer from over-simplified posterior approximations, the adversarial autoencoder (AAE) has shown promise by adopting GAN to match the variational posterior to an arbitrary prior through adversarial training. Both VAEs and GANs face significant challenges such as training stability, mode collapse, and difficulty in extracting meaningful latent representations. In this paper, we propose the Multi-adversarial Autoencoder (MAAE), which extends the AAE framework by incorporating multiple discriminators and enabling soft-ensemble feedback. By adaptively regulating the collective feedback from multiple discriminators, MAAE captures a balance between fitting the data distribution and performing accurate inference and accelerates training stability while extracting meaningful and interpretable latent representations. Experimental evaluations on MNIST, CIFAR10, and CelebA datasets demonstrate significant improvements in latent representation, quality of generated samples, log-likelihood, and a pairwise comparison metric, with comparisons to recent methods.

Spectrum-guided GAN with density-directionality sampling: Diverse high-fidelity signal generation for fault diagnosis of rotating machinery

In the field of fault diagnosis for rotating machinery, where available fault data are limited, numerous studies have employed a generative adversarial network (GAN) for data generation. However, the limited fault data for training GAN exacerbate GAN’s inherent training instability and mode collapse issues, which are induced by adversarial training. Moreover, the stochastic nature of random sampling for latent vectors sampling often results in low-fidelity and poor diversity generation, which negatively affects the fault diagnosis models. To address these issues, this paper presents two novel approaches: a spectrum-guided GAN (SGAN) and density-directionality sampling (DDS). SGAN mitigates training instability and mode collapse through combinatorial data utilization, adversarial spectral loss, and a tailored model structure. DDS ensures the high-fidelity and high-diversity of the generated data by selectively sampling the latent vectors through two steps: density-based filtering and directionality-based sampling in the feature space. Validation on both rotor and rolling element bearing datasets demonstrates that SGAN-DDS considerably improves classification results under the limited fault data. Furthermore, fidelity and diversity analyses are conducted to validate DDS, which increase the credibility of the proposed method; and offer advancement toward the application of deep-learning and GAN in industrial fields.

Closing the Gap Between Theory and Practice During Alternating Optimization for GANs.

Synthesizing high-quality and diverse samples is the main goal of generative models. Despite recent great progress in generative adversarial networks (GANs), mode collapse is still an open problem, and mitigating it will benefit the generator to better capture the target data distribution. This article rethinks alternating optimization in GANs, which is a classic approach to training GANs in practice. We find that the theory presented in the original GANs does not accommodate this practical solution. Under the alternating optimization manner, the vanilla loss function provides an inappropriate objective for the generator. This objective forces the generator to produce the output with the highest discriminative probability of the discriminator, which leads to mode collapse in GANs. To address this problem, we introduce a novel loss function for the generator to adapt to the alternating optimization nature. When updating the generator by the proposed loss function, the reverse Kullback-Leibler divergence between the model distribution and the target distribution is theoretically optimized, which encourages the model to learn the target distribution. The results of extensive experiments demonstrate that our approach can consistently boost model performance on various datasets and network structures.

Advanced 3D Face Reconstruction from Single 2D Images Using Enhanced Adversarial Neural Networks and Graph Neural Networks.

This paper presents a novel framework for 3D face reconstruction from single 2D images and addresses critical limitations in existing methods. Our approach integrates modified adversarial neural networks with graph neural networks to achieve state-of-the-art performance. Key innovations include (1) a generator architecture based on Graph Convolutional Networks (GCNs) with a novel loss function and identity blocks, mitigating mode collapse and instability; (2) the integration of facial landmarks and a non-parametric efficient-net decoder for enhanced feature capture; and (3) a lightweight GCN-based discriminator for improved accuracy and stability. Evaluated on the 300W-LP and AFLW2000-3D datasets, our method outperforms existing approaches, reducing Chamfer Distance by 62.7% and Earth Mover's Distance by 57.1% on 300W-LP. Moreover, our framework demonstrates superior robustness to variations in head positioning, occlusion, noise, and lighting conditions while achieving significantly faster processing times.

The effect of cavity shape on critical high-temperature regions formation pattern and structural evolution difference in explosive particle bed

The occurrence of critical high-temperature regions within explosives is a crucial factor in initiating combustion and explosion. The primary causes of hot regions in cavity-containing explosives are dissipative processes occurring inside the explosive or at its interface, such as shear, friction, viscosity, plasticity, and heat transfer. Currently, there is limited research on the generation and dissipation mechanisms of critical high-temperature regions in cavity-containing explosives during impacts, based on the mesoscale analysis that describes particle motion. In this study, a model based on the discrete element method is employed allowing for an accurate analysis of the dynamic and thermodynamic processes among particles, which systematically considers the shear history between particles, elastoplastic collisions, and heat transfer between particles. The temporal evolution of particle temperature and dissipated work contours during the impact process of explosives with different cavity shapes is analyzed. Explosives with cavities undergo three stages during the impact process: the formation of the trapezoidal high-temperature region, cavity collapse, and dispersion of the particle bed, ultimately leading to the formation of two main critical high-temperature regions: near the cavity and near the wall. The temperature rise of hot particles near the wall can be roughly divided into two stages, and the temperature rise mechanisms are different. Initially, the temperature rise is primarily attributed to plastic dissipation of hot particles, and the same initial impact velocity results in the same temperature rise. During the cavity collapsing, continuous momentum transfer occurs between the hot particles near the wall and the particles in the high-temperature shear bands, deeply influencing the development of the high-temperature shear bands through the movement of the hot particles near the cavity. Therefore, the different collapse modes of hot particles near cavities of different shapes lead to differences in the later temperature evolution of the hot particles near the wall. The unique aggregation processes of hot particles near cavities of various shapes result in distinct high-temperature region morphologies: circular cavity exhibits approximate “——” shape, triangular cavity displays arch shape, and inverted triangular cavity presents “M” shape. Significantly, a new mechanism resulting from the interaction of two opposing “jet “flows for the formation of higher-temperature hot regions near the circular cavity has been discovered.