Search

In recent years, the development of deep learning has further improved hash retrieval technology. Most of the existing hashing methods currently use Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) to process image and text information, respectively. This makes images or texts subject to local constraints, and inherent label matching cannot capture finegrained information, often leading to suboptimal results. Driven by the development of the transformer model, we propose a framework called ViT2CMH mainly based on the Vision Transformer to handle deep Cross-modal Hashing tasks rather than CNNs or RNNs. Specifically, we use a BERT network to extract text features and use the vision transformer as the image network of the model. Finally, the features are transformed into hash codes for efficient and fast retrieval. We conduct extensive experiments on Microsoft COCO (MS-COCO) and Flickr30K, comparing with baselines of some hashing methods and image-text matching methods, showing that our method has better performance. [ABSTRACT FROM AUTHOR]

As the main entertainment place for human beings, the hidden problems of the Internet are becoming more and more obvious, of which cyber-violence is one of the most typical problems at present. When studying the factors that influence cyber violence, various technologies and methods are used to explain the relationships between specific variables. Unexpectedly, the structural equation model (SEM) has acquired very little concern in factors affecting cyber-violence incidents. In order to solve the limitations of this knowledge body, this study is to establish a model by using structural equation modeling to determine the key factors affecting college students' cyber-violence. In this paper, by designing a questionnaire to obtain relevant data, five potential influencing factors of cyber violence are quantified, including netizens' behavior factors, moral sentiment factors, netizens' psychological factors, social environment factors and behavior analysis factors. On the basis of excavating the multi-layer complex interactive relationship between the influencing factors of cyber violence, ten hypothetical relationships are established and compared by using a structural equation model. Through the analysis of the influencing factors of cyber-violence, it is recognized that reasonably guiding cyber groups is one of the important way to solve extreme mass incidents. [ABSTRACT FROM AUTHOR]