3D Range-Gated Imaging technology is widely used for detection in complex environments (such as autonomous driving scenarios) due to its excellent anti-interference capabilities. However, its application faces the dual challenges of a lack of specialized datasets and the limited performance of traditional RGB models in low signal-to-noise ratio environments, which hinders the transfer and generalization of deep learning methods. To address these difficulties, this paper proposes a 3D imaging method based on multimodal feature fusion. Specifically, the model adopts a dual Vision Transformer (ViT) encoder, single-decoder architecture. On one hand, it performs pre-trained ViT encoding on geometrically re-projected RGB images. On the other hand, it applies an isomorphic ViT encoding to the range-gated images. Through layer-wise semantic recombination, it achieves efficient cross-modal feature fusion, not only does it enhance the robustness and accuracy of depth estimation, but it can also be easily deployed on edge devices. To overcome the problem of overfitting to LiDAR ground truth data, a spatially constrained window cropping data augmentation strategy is designed, significantly increasing the diversity of training samples and the model's generalization ability. To address the input resolution limitations of Transformers, an optimization scheme combining dynamic patch-based training and progressive up-sampling is further proposed, balancing high-resolution feature representation with efficient training. Experimental results show that the proposed method reduces the depth estimation RMSE on a public test set by more than 12% compared to mainstream baseline models, with particularly outstanding performance in low-texture and long-distance scenes. This research provides a systematic technical solution for cross-modal 3D perception and offers theoretical and engineering references for designing 3D imaging models for complex environments.

Efficient resource allocation is pivotal for achieving high performance in emerging computer systems, where multiple users and tasks compete for shared resources. This challenge spans various domains, including data centers, multicore processors, cloud computing and edge computing, each requiring nuanced allocation strategies to balance competing demands. Traditional approaches often assume concave utility (performance) functions for users, simplifying optimization but failing to capture the complexities of real-world scenarios where non-concave utility functions prevail. Numerous works in the literature apply the greedy algorithm to nonconcave utility functions, resulting in suboptimal solution due to the short-sighted behaviors. To improve this gap, we propose a novel multiphase resource allocation framework that accurately reflects the non-linear dynamics of these systems. To tackle the NP-complete nature of this problem, we formulate a customized resource allocation Markov Decision Process (MDP) that integrates the characteristics of multi-phase utility functions into a nuanced design of the key MDP components, such as state representations, reward signals, and actions. We explore two reinforcement learning (RL)-based methods, specifically Dueling Deep Q-Network (Dueling DQN) and Proximal Policy Optimization (PPO), to optimize resource allocation over time. Our RL-based strategies outperform the conventional greedy algorithm by approximately 37% in standard environments and up to 73% in specialized environments, highlighting their effectiveness in handling the resource allocation problem with non-concave utility functions and achieving scalable, real-time solutions.

Abstract Existing crowd evacuation simulation methods commonly face challenges of low efficiency in path planning and insufficient realism in pedestrian movement during the evacuation process. In this study, we propose a novel crowd evacuation path planning approach based on the learning curve–deep deterministic policy gradient (LC‐DDPG) algorithm. The algorithm incorporates dynamic experience pool and a priority experience sampling strategy, enhancing convergence speed and achieving higher average rewards, thus efficiently enabling global path planning. Building upon this foundation, we introduce a double‐layer method for crowd evacuation using deep reinforcement learning. Specifically, within each group, individuals are categorized into leaders and followers. At the top layer, we employ the LC‐DDPG algorithm to perform global path planning for the leaders. Simultaneously, at the bottom layer, an enhanced social force model guides the followers to avoid obstacles and follow the leaders during evacuation. We implemented a crowd evacuation simulation platform. Experimental results show that our proposed method has high path planning efficiency and can generate more realistic pedestrian trajectories in different scenarios and crowd sizes.

Abstract Background Cartoon images play a vital role in film production, scientific and educational animation, video games, and other fields, and are one of the key visual expressions of artistic creation. However, since hand‐crafted cartoon images often require a great deal of time and effort on the part of professional artists, it is necessary to be able to automatically transform real‐world images into different styles of cartoon images. Although cartoon images vary from artist to artist, cartoon images generally have the unique characteristics of being highly simplified and abstract, with clear edges, smooth color shading, and relatively simple textures. However, existing image cartoonization methods tend to create a number of problems when performing style transfer, which mainly include: (1) the resulting generated images do not have obvious cartoon‐style textures; and (2) the generated images are prone to structural confusion, color artifacts, and loss of the original image content. Therefore, it is also a great challenge in the field of image cartoonization to be able to make a good balance between style transfer and content keeping. Methods In this paper, we propose a GAN‐based multi‐attention mechanism for image cartoonization to address the above issues. The method combines the residual block used to extract deep network features in the generator with the attention mechanism, and further strengthens the perceptual ability of the generative model to cartoon images through the adaptive feature correction of the attention module to improve the cartoon features of the generated images. At the same time, we also introduce the attention mechanism in the convolution block of the discriminator, which is used to further reduce the image visual quality problem caused by the style transfer process. By introducing the attention mechanism into the generator and discriminator models of the generative adversarial network, our method enables the generated images to have obvious cartoon‐style features while effectively improving the image's visual quality. Results A large number of quantitative, qualitative, and ablation experiments are conducted to demonstrate the advantages of our method in the field of image cartoonization and the role of each module in the method.

Word-text matrix has been usually used as text representation model in text classification and text clustering. However its high dimension and sparsity reduce its expression ability. For improving its expression ability, authors mine word word relation and text-text relation, and integrate these semantic relationships into word-text matrix. The classification experiments show that these new representation models can improve the classification accuracy of text efficiently as well as represent the text information better.

Aspect mining improves the modularity of legacy software systems through identifying their underlying crosscutting concerns (CCs). However, a realistic CC is a composite one that consists of CC seeds and relative program elements, which makes it a great challenge to identify a composite CC. In this paper, inspired by the state-of-the-art information retrieval techniques, we model this problem as a semi-supervised learning problem. First, the link analysis technique is adopted to generate CC seeds. Second, we construct a coupling graph, which indicates the relationship between CC seeds. Then, we adopt community detection technique to generate groups of CC seeds as constraints for semi-supervised learning, which can guide the clustering process. Furthermore, we propose a semi-supervised graph clustering approach named constrained authority-shift clustering to identify composite CCs. Two measurements, namely, similarity and connectivity, are defined and seeded graph is generated for clustering program elements. We evaluate constrained authority-shift clustering on numerous software systems including large-scale distributed software system. The experimental results demonstrate that our semi-supervised learning is more effective in detecting composite CCs. Copyright © 2013 John Wiley & Sons, Ltd.

Software architecture recovery is to gain the architectural level understanding of a software system while its architecture description does not exist. In recent years, researchers have adopted various software clustering techniques to detect hierarchical structure of software systems. Most graph clustering techniques focus on the connectivity between program elements, but unreasonably ignore the similarity which is also a key measure for finding elements of one module. In this paper we propose a novel hierarchy graph clustering algorithm DGHC, which considers both similarity and connectivity between program elements. During the transformation of program dependence graph edges representing similarity between elements are added. Then similar elements are grouped by density-based approaches. The alternative strategy is adopted to find groups of closely connected and similar elements. Meanwhile we adjust the contribution of connectivity and similarity by a flexible clustering algorithm based on short random walk model, which can obtain more structure information of software to find its multiple layers. Furthermore a new method called Multi-layer Propagation Gap is proposed to suggest stable layers of hierarchy clustering result as multiple layers of software system. Extensive experimental results illustrate the effectiveness and efficiency of DGHC in detecting hierarchy structure of software through comparison with various software clustering methods.