Virtual human is widely employed in various industries, including personal assistance, intelligent customer service, and online education, thanks to the rapid development of artificial intelligence. An anthropomorphic digital human can quickly contact people and enhance user experience in human–computer interaction. Hence, we design the human–computer interaction system framework, which includes speech recognition, text-to-speech, dialogue systems, and virtual human generation. Next, we classify the model of talking-head video generation by the virtual human deep generation framework. Meanwhile, we systematically review the past five years’ worth of technological advancements and trends in talking-head video generation, highlight the critical works and summarize the dataset.

Named entity recognition (NER) is a subfield of natural language processing (NLP) that identifies and classifies entities from plain text, such as people, organizations, locations, and other types. NER is a fundamental task in information extraction, information retrieval, and text summarization, as it helps to organize the relevant information in a structured way. The current approaches to Chinese named entity recognition do not consider the category information of matched Chinese words, which limits their ability to capture the correlation between words. This makes Chinese NER more challenging than English NER, which already has well-defined word boundaries. To improve Chinese NER, it is necessary to develop new approaches that take into account category features of matched Chinese words, and the category information would help to effectively capture the relationship between words. This paper proposes a Prompt-based Word-level Information Injection BERT (PWII-BERT) to integrate prompt-guided lexicon information into a pre-trained language model. Specifically, we engineer a Word-level Information Injection Adapter (WIIA) through the original Transformer encoder and prompt-guided Transformer layers. Thus, the ability of PWII-BERT to explicitly obtain fine-grained character-to-word relevant information according to the category prompt is one of its key advantages. In experiments on four benchmark datasets, PWII-BERT outperforms the baselines, demonstrating the significance of fully utilizing the advantages of fusing the category information and lexicon feature to implement Chinese NER.

Information diffusion prediction aims to predict the next infected user in the information diffusion, which is a critical task to understand how information spreads on social platforms. Existing methods mainly focus on the sequences or topology structure in euclidean space. However, they fail to sufficiently consider the hierarchical structure or power-law structure of the underlying topology of information cascade graphs and social networks, resulting in distortion of user features. To tackle above issue, we propose an innovative Constrained Temporal Hypergraphs and Graph Neural Networks (THGNets) framework that is tailored for information diffusion prediction. Specifically, we introduce hyperbolic temporal hypergraphs neural network to alleviate the distortion of user features by hyperbolic hierarchical learning in information cascades. Additionally, it also captures high-order dynamic interaction patterns between users and further integrates the time-consistency constraint mechanism to mitigate the instability and non-smoothness of user features in latent space. In parallel, we apply the hyperbolic graph neural network to investigate the hierarchical structure and user homogeneity on social networks, enhancing our understanding of social relationships. Moreover, hyperbolic gated recurrent units are employed to capture the potential dependency relationships between contextual users. Experiments conducted on four public datasets demonstrate that the proposed THGNets significantly outperform the existing methods, thereby validating the superiority and rationality of our approach.

Audio-driven cross-modal talking head generation has experienced significant advancement in the last several years, and it aims to generate a talking head video that corresponds to a given audio sequence. Out of these approaches, the NeRF-based method can generate videos featuring a specific person with more natural motion compared to the one-shot methods. However, previous approaches failed to distinguish the importance of different regions, resulting in the loss of information-rich region features. To alleviate the problem and improve video quality, we propose MLDF-NeRF, an end-to-end method for talking head generation, which can achieve better vector representation through multi-level feature dynamic fusion. Specifically, we designed two modules in MLDF-NeRF to enhance the cross-modal mapping ability between audio and different facial regions. We initially developed a multi-level tri-plane hash representation that uses three sets of tri-plane hash networks with varying resolutions of limitation to capture the dynamic information of the face more accurately. Then, we introduce the idea of multi-head attention and design an efficient audio-visual fusion module that explicitly fuses audio features with image features from different planes, thereby improving the mapping between audio features and spatial information. Meanwhile, the design helps to minimize interference from facial areas unrelated to audio, thereby improving the overall quality of the representation. The quantitative and qualitative results indicate that our proposed method can effectively generate talk heads with natural actions and realistic details. Compared with previous methods, it performs better in terms of image quality, lip sync, and other aspects.