Multimodal personality perception method and device based on multiple correlation features and graph relationship attention

Abstract

The present invention relates to the field of computer vision, and more particularly to a multimodal personality perception method and apparatus that utilizes multiple association features and graph relationship attention. The method comprises: obtaining an input video for personality perception; inputting the input video into a data preprocessing module to obtain visual modality input, audio modality input, and text modality input; inputting the visual modality input, audio modality input, and text modality input into a modal feature extraction network module to obtain scene-audio association features, scene-description word association features, audio-description word association features, and text modality features; inputting the scene-audio association features, scene-description word association features, audio-description word association features, and text modality features into a feature fusion module to obtain multimodal fusion features; and inputting the multimodal fusion features into a perception prediction module to obtain personality perception results. The present invention proposes a multimodal attention fusion framework for personality perception.

Description

Technical Field

[0001] The present invention relates to the field of computer vision technology, and in particular to a multimodal personality perception method and device that pays attention to multiple associated features and graph relationships. Background Art

[0002] Personality generally encompasses aspects of behavior, temperament, emotions, and psychology that define an individual's unique character. Explicit personality perception refers to the perception of a person based on their observed behavior and appearance, which differs significantly from actual personality (representing a person's true psychological characteristics). Personality perception aims to identify implicit personality traits in social media posts, providing insights into human behavior, emotional processes, and mental health. Personality perception helps people better understand themselves and improve their mental health, and plays an important role in fields such as psychotherapy, clinical diagnosis, career development, and social science research. In personality research, researchers have developed psychological scales and models to understand and measure personality traits. For example, the Big Five personality model measures personality through five traits: openness (O), conscientiousness (C), extraversion (E), agreeableness (A), and neuroticism (N). As an essential component of social interaction, explicit personality perception significantly influences how people respond to others. Therefore, it has important applications in interpersonal interaction analysis, human-computer interaction systems, social robotics, education, marketing, and recommendation systems.

[0003] With the rise of multimedia social platforms like YouTube and TikTok, people are increasingly using videos to share their lives. To address this trend, the European Conference on Computer Vision (ECCV) and the Conference on Computer Vision and Pattern Recognition (CVPR) held competitions on personality trait perception in 2016 and 2017, generating the second edition of the impression dataset. As a result, automatically perceiving personality from videos has become more feasible. In recent years, pre-training has achieved significant success in multimodal processing. For example, pre-trained VL (Vision-Language) models have demonstrated superior performance and scalability in both understanding tasks (such as text-visual retrieval) and generation tasks (such as video captioning). Pre-trained audio models are capable of representing complex audio information. Learning general correlations between vision, text, and audio from scratch, such as with the VATT (Video-Audio-Text Transformer), is computationally intensive. Therefore, an intuitive idea is to combine state-of-the-art vision-language models with pre-trained audio backbones. With the outstanding performance of multimodal large-scale pre-trained models in various visual downstream tasks and the emergence of graph structure learning methods, introducing them into personality perception tasks will surely bring unexpected results.

[0004] According to some studies, language and unconscious behavior, including audio (voice tone, acoustic intensity), visual cues (facial expressions, eye contact, body gestures), text (audio transcriptions), and contextual information, are important components of the personality perception system. These multimodal features contribute significantly to the practical assessment of personality and emotion. By leveraging the complementarity of information from different modalities, multimodal personality perception methods can significantly enhance the robustness of perception. Therefore, various methods based on different modalities have been developed to predict personality traits. Despite the achievements of these research methods, they still face two key challenges:

[0005] 1. Most existing methods only focus on extracting more single-modal features, ignoring the importance of correlation features between different modalities.

[0006] 2. Although there are many multimodal feature fusion methods, their actual effects still have room for improvement. Summary of the Invention

[0007] To address the technical issues of existing technologies that ignore the correlation features between different modalities and the characteristics of graph relational attention networks in extracting key temporal information, the present invention provides a multimodal personality perception method and device based on multiple correlation features and graph relational attention. The technical solution is as follows:

[0008] In one aspect, a multimodal personality perception method based on multiple correlation features and graph relationship attention is provided. The method is implemented by a multimodal personality perception device and includes:

[0009] S1. Obtain an input video for personality perception.

[0010] S2. Input the input video to the data preprocessing module to obtain visual modality input, audio modality input, and text modality input.

[0011] S3. Input the visual modal input, audio modal input and text modal input into the modal feature extraction network module to obtain scene-audio association features, scene-description word association features, audio-description word association features and text modal features.

[0012] S4. Input the scene-audio association features, scene-description word association features, audio-description word association features and text modality features into the feature fusion module to obtain multimodal fusion features.

[0013] S5. Input the multimodal fusion features into the perception prediction module to obtain the personality perception results.

[0014] Optionally, in S3, the visual modality input, the audio modality input, and the text modality input are input into the modality feature extraction network module to obtain scene-audio association features, scene-description word association features, audio-description word association features, and text modality features, including:

[0015] S31. Input the visual modality input and the audio modality input into a scene-audio correlation feature extraction module to obtain scene-audio correlation features.

[0016] S32: Input the visual modality input into the scene-description word association feature extraction module to obtain the scene-description word association feature.

[0017] S33. Input the audio modal input into the audio-description word association feature extraction module to obtain the audio-description word association feature.

[0018] S34. Input the text modal input into the Roberta model to obtain text modal features.

[0019] Optionally, in S31, the visual modality input and the audio modality input are input into a scene-audio correlation feature extraction module to obtain scene-audio correlation features, including:

[0020] S311. For visual modality input, use a pre-trained image encoder to extract scene image features, and construct a visual relationship graph based on the scene image features.

[0021] S312: Preprocess the audio modal input to obtain a plurality of mel-spectrogram images, extract a plurality of audio features from the plurality of mel-spectrogram images using a pre-trained audio encoder, and construct an audio relationship graph based on the plurality of audio features.

[0022] S313. Design a graph relationship attention network, obtain visual relationship graph features based on the visual relationship graph and the graph relationship attention network; obtain audio relationship graph features based on the audio relationship graph and the graph relationship attention network.

[0023] S314: Concatenate the visual relationship graph features and the audio relationship graph features to obtain scene-audio association features.

[0024] Optionally, the step of inputting the visual modality input into the scene-description word association feature extraction module in S32 to obtain the scene-description word association feature includes:

[0025] S321. Construct a list of personality descriptive words.

[0026] S322: Generate personality description word embedding based on the personality description word list and the text encoder.

[0027] S323: Generate scene image embedding based on the visual modality input and the image encoder.

[0028] S324. Calculate the dot product of the personality description word embedding and the scene image embedding to obtain the scene-description word association feature.

[0029] Optionally, the step of inputting the audio modal input into the audio-description word association feature extraction module in S33 to obtain the audio-description word association features includes:

[0030] S331. Generate personality description word embedding based on the personality description word list and the text encoder.

[0031] S332: Generate audio embedding based on the audio modality input and the pre-trained audio encoder.

[0032] S333. Calculate the dot product of the personality description word embedding and the audio embedding to obtain the audio-description word association feature.

[0033] Optionally, in S4, the scene-audio association feature, the scene-description word association feature, the audio-description word association feature, and the text modality feature are input into a feature fusion module to obtain a multimodal fusion feature, including:

[0034] S41. Calculate the scene-audio association features, scene-description word association features, audio-description word association features and text modality features through the first attention score calculation module, the second attention score calculation module, the third attention score calculation module and the fourth attention score calculation module, respectively, to obtain the scene-audio association feature self-attention score, the scene-description word association feature self-attention score, the audio-description word association feature self-attention score and the text modality feature self-attention score.

[0035] S42. Based on the scene-audio associated feature self-attention score, the scene-description word associated feature self-attention score, the audio-description word associated feature self-attention score, and the text modality feature self-attention score, calculate the normalized value of the scene-audio associated feature self-attention score, the normalized value of the scene-description word associated feature self-attention score, the normalized value of the audio-description word associated feature self-attention score, and the normalized value of the text modality feature self-attention score.

[0036] S43. Calculate the multimodal fusion feature based on the normalized value of the scene-audio associated feature self-attention score, the normalized value of the scene-description word associated feature self-attention score, the normalized value of the audio-description word associated feature self-attention score, and the normalized value of the text modality feature self-attention score.

[0037] On the other hand, a multimodal personality perception device based on multiple correlation features and graph relationship attention is provided. The device is applied to a multimodal personality perception method based on multiple correlation features and graph relationship attention. The device includes:

[0038] The acquisition module is used to obtain the input video to be used for personality perception.

[0039] The data preprocessing module is used to input the input video into the data preprocessing module to obtain visual modality input, audio modality input and text modality input.

[0040] The modal feature extraction network module is used to input visual modal input, audio modal input and text modal input into the modal feature extraction network module to obtain scene-audio association features, scene-description word association features, audio-description word association features and text modal features.

[0041] The feature fusion module is used to input scene-audio association features, scene-description word association features, audio-description word association features and text modality features into the feature fusion module to obtain multimodal fusion features.

[0042] The output module is used to input the multimodal fusion features into the perception prediction module to obtain the personality perception results.

[0043] Optionally, the modal feature extraction network module is further used to:

[0044] S31. Input the visual modality input and the audio modality input into a scene-audio correlation feature extraction module to obtain scene-audio correlation features.

[0045] S32: Input the visual modality input into the scene-description word association feature extraction module to obtain the scene-description word association feature.

[0046] S33. Input the audio modal input into the audio-description word association feature extraction module to obtain the audio-description word association feature.

[0047] S34. Input the text modal input into the Roberta model to obtain text modal features.

[0048] Optionally, the modal feature extraction network module is further used to:

[0049] S311. For visual modality input, use a pre-trained image encoder to extract scene image features, and construct a visual relationship graph based on the scene image features.

[0050] S312: Preprocess the audio modal input to obtain a plurality of mel-spectrogram images, extract a plurality of audio features from the plurality of mel-spectrogram images using a pre-trained audio encoder, and construct an audio relationship graph based on the plurality of audio features.

[0051] S313. Design a graph relationship attention network, obtain visual relationship graph features based on the visual relationship graph and the graph relationship attention network; obtain audio relationship graph features based on the audio relationship graph and the graph relationship attention network.

[0052] S314: Concatenate the visual relationship graph features and the audio relationship graph features to obtain scene-audio association features.

[0053] Optionally, the modal feature extraction network module is further used to:

[0054] S321. Construct a list of personality descriptive words.

[0055] S322: Generate personality description word embedding based on the personality description word list and the text encoder.

[0056] S323: Generate scene image embedding based on the visual modality input and the image encoder.

[0057] S324. Calculate the dot product of the personality description word embedding and the scene image embedding to obtain the scene-description word association feature.

[0058] Optionally, the modal feature extraction network module is further used to:

[0059] S331. Generate personality description word embedding based on the personality description word list and the text encoder.

[0060] S332: Generate audio embedding based on the audio modality input and the pre-trained audio encoder.

[0061] S333. Calculate the dot product of the personality description word embedding and the audio embedding to obtain the audio-description word association feature.

[0062] Optionally, the feature fusion module is further configured to:

[0063] S41. Calculate the scene-audio association features, scene-description word association features, audio-description word association features and text modality features through the first attention score calculation module, the second attention score calculation module, the third attention score calculation module and the fourth attention score calculation module, respectively, to obtain the scene-audio association feature self-attention score, the scene-description word association feature self-attention score, the audio-description word association feature self-attention score and the text modality feature self-attention score.

[0064] S42. Based on the scene-audio associated feature self-attention score, the scene-description word associated feature self-attention score, the audio-description word associated feature self-attention score, and the text modality feature self-attention score, calculate the normalized value of the scene-audio associated feature self-attention score, the normalized value of the scene-description word associated feature self-attention score, the normalized value of the audio-description word associated feature self-attention score, and the normalized value of the text modality feature self-attention score.

[0065] S43. Calculate the multimodal fusion feature based on the normalized value of the scene-audio associated feature self-attention score, the normalized value of the scene-description word associated feature self-attention score, the normalized value of the audio-description word associated feature self-attention score, and the normalized value of the text modality feature self-attention score.

[0066] In another aspect, a multimodal personality perception device is provided. The multimodal personality perception device includes: a processor; and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, implement any one of the multimodal personality perception methods using multiple association features and graph relationship attention as described above.

[0067] On the other hand, a computer-readable storage medium is provided, wherein the storage medium stores at least one instruction, and the at least one instruction is loaded and executed by a processor to implement any one of the above-mentioned multimodal personality perception methods based on multiple association features and graph relationship attention.

[0068] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0069] In an embodiment of the present invention, a multimodal attention fusion framework is proposed for personality perception. The framework utilizes multiple correlation features and graph relation attention networks and is designed specifically for vision-language-audio processing. The present invention includes audio modalities in a unified manner by extending the vision-language pre-training model CLIP. The present invention utilizes contrastive learning techniques from the perspectives of cross-modality and intra-modality to fully utilize the multimodal pre-training model. In order to extract scene-audio correlation features, a scene-audio correlation feature extraction method is proposed in the framework of the present invention. In addition, the present invention extends the CLIP model to CLIPAudio audio, which is used to mine the correlation features of scene personality descriptors and audio personality descriptions respectively. The present invention also utilizes the pre-trained Roberta model to extract text-specific features. In order to effectively fuse these multimodal information, the present invention designs a multimodal attention feature fusion strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0071] Figure 1 This is a flow chart of a multimodal personality perception method based on multiple correlation features and graph relationship attention provided by an embodiment of the present invention;

[0072] Figure 2 This is an overall architecture diagram provided by an embodiment of the present invention;

[0073] Figure 3 is a flow chart of a scene audio correlation feature extraction method provided by an embodiment of the present invention;

[0074] Figure 4 It is a graph relationship attention network graph provided by an embodiment of the present invention;

[0075] Figure 5 This is a flow chart of a method for extracting associated features of scene and audio personality descriptions provided by an embodiment of the present invention;

[0076] Figure 6This is a flow chart of the text embedding feature extraction method based on Roberta provided in an embodiment of the present invention;

[0077] Figure 7 This is a flow chart of a multimodal attention feature fusion strategy provided by an embodiment of the present invention;

[0078] Figure 8 This is a block diagram of a multimodal personality perception device that pays attention to multiple associated features and graph relationships provided by an embodiment of the present invention;

[0079] Figure 9 This is a schematic diagram of the structure of a multimodal personality perception device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0080] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0081] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0082] In the embodiments of the present invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same. The terms "of," "corresponding," and "corresponding" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same.

[0083] In the embodiments of the present invention, sometimes a subscript such as W1 may be written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0084] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0085] The embodiment of the present invention provides a multimodal personality perception method based on multiple associated features and graph relationship attention. The method can be implemented by a multimodal personality perception device, which can be a terminal or a server. Figure 1 The flowchart of the multimodal personality perception method using multiple correlation features and graph relationship attention is shown. The processing flow of the method may include the following steps:

[0086] S1. Obtain an input video for personality perception.

[0087] S2. Input the input video to the data preprocessing module to obtain visual modality input, audio modality input, and text modality input.

[0088] S3. Input the visual modal input, audio modal input and text modal input into the modal feature extraction network module to obtain scene-audio association features, scene-description word association features, audio-description word association features and text modal features.

[0089] S4. Input the scene-audio association features, scene-description word association features, audio-description word association features and text modality features into the feature fusion module to obtain multimodal fusion features.

[0090] S5. Input the multimodal fusion features into the perception prediction module to obtain the personality perception results.

[0091] To address the challenges of existing technologies, this paper proposes a multimodal personality perception framework (CLIP for Vision, Language, and Audio). This framework extends CLIP (Contrastive Language-Image Pre-Training) to accommodate the audio modality and employs a unified triple encoder architecture for multimodal processing. This framework integrates various methods for learning high-level multimodal feature representations. Specifically, this paper first employs CLIP, a leading vision-language model, as the visual and text encoders. To ensure training consistency and efficiency, this paper uses an audio encoder with a similar architecture to the visual encoder. This paper proposes a method for extracting scene-audio correlation features. To simultaneously encode visual scenes, audio information, and personality descriptors, this paper designs the CLIPAudio method to extract scene-personality descriptor correlation and audio-personality descriptor correlation features. To better utilize the multimodal representations learned by this framework, this paper further employs the pretrained Roberta model to extract text features. Furthermore, to better focus on important features and reduce redundant ones, we designed a multimodal attention feature fusion strategy to fuse multiple features and train an MLP (Multilayer Perceptron) model to predict personality traits. For efficient modeling, we employed a two-stage training process. First, we trained each modality-specific network separately. Then, we combined these sub-networks and fine-tuned the entire framework in a multimodal manner.

[0092] Furthermore, GNNs (Graph Neural Networks) have recently been widely used in various challenging tasks due to their ability to model data generated in non-Euclidean domains and capture internal dependencies within the data. The key advantage of GCNs (Graph Convolution Networks) is that they can directly learn latent embeddings for nodes based on their neighbors and their relationships, achieving widespread success. They have been extensively studied for exploring relational data structures, contextual information, and geometric dependencies. Similar to how convolution operations in CNNs (Convolutional Neural Networks) aggregate information from neighboring pixels within a predefined grid cell, graph convolutions aggregate information from neighboring vertices within a predefined graph distance. In addition to applying GCNs to landmark-based facial data for FER (Facial Expression Recognition), a significant body of research has also applied GCNs to exploring correlations between different frames in a video for video-based FER, as well as correlations between different facial AUs (Action Units) for FER. However, the above works have not yet considered the attention mechanism in GCNs and CNN-based learning, which can model scene images or log-mel spectrograms as images for personality prediction. Therefore, this paper proposes a graph relation attention network based on CLIP's ViT encoder and VGG-like CNN for personality prediction.

[0093] Conceptually, the CLIP model consists of two subnetworks: a text encoder and an image encoder. These two components of the CLIP model are jointly pre-trained under natural language supervision. This training setup enables the model to generalize its classification capabilities to image samples from previously unseen datasets, based on the provided labels, without requiring additional fine-tuning. For the text encoding component, a slightly modified Transformer architecture was employed. For the chosen 12-layer model, the input text was encoded using lowercase byte pairs with a vocabulary size of 49,408. Due to computational limitations, the maximum sequence length was limited to 76. Two different architectures were considered for the image encoding component of the CLIP model. One, the Vision Transformer (ViT), has an architecture similar to a text header. Multimodal designs were first applied to video-related tasks and later adapted to sound classification. However, despite their multimodal design, these methods have at most used two modalities simultaneously, while recent research has shown the benefits of using more modalities. These multimodal methods share a key idea from contrastive learning, a branch of self-supervised learning that helps overcome the lack of high-quality labeled data. Therefore, the present invention proposes an extension of the CLIP model that can process audio data in addition to text and images.

[0094] like Figure 2 As shown in Figure 2, the framework proposed in this paper aims to predict individual personality traits based on short videos. The framework includes four main steps: (1) data preprocessing; (2) modal feature extraction network; (3) feature fusion and model training; (4) perception prediction. Specifically:

[0095] S1. Obtain an input video for personality perception.

[0096] S2. Input the input video to the data preprocessing module to obtain visual modality input, audio modality input, and text modality input.

[0097] In one feasible implementation, during the data preprocessing stage, the input video is segmented into multiple modal inputs, including visual, audio, and text. The input short video covers three natural modalities: the visual modality is used for image sequences, the audio modality is used for speech, and the text modality is used for transcripts and personality descriptors. During the feature extraction stage, the framework of the present invention uses modality-specific networks to learn personality trait information from each input modality separately. Different modality-specific features are obtained through the respective specifically learned networks. These features are fused through the multimodal attention fusion strategy to fuse the different modal features, and then input into the regression head to predict the final scores of various personality traits.

[0098] S3. Input the visual modal input, audio modal input and text modal input into the modal feature extraction network module to obtain scene-audio association features, scene-description word association features, audio-description word association features and text modal features.

[0099] In one feasible implementation, step S3 above mainly extracts features from three types of multimodal correlation features using the method proposed in the present invention. These are: using a method for extracting scene-audio correlation features, and using the CLIPAudio method to extract scene-description word correlation features and audio-description word correlation features. Next, the present invention will elaborate on these feature extraction methods as follows:

[0100] S31. Input the visual modality input and the audio modality input into a scene-audio correlation feature extraction module to obtain scene-audio correlation features.

[0101] Figure 3 The framework of the scene-audio correlation feature extraction method based on the present invention is described, which consists of three parts: 1) construction of visual and audio relationship graph; 2) graph relationship attention network; 3) extraction of scene-audio correlation features.

[0102] Specifically, the above step S31 may include the following steps S311-S314:

[0103] S311. For visual modality input, use a pre-trained image encoder to extract scene image features, and construct a visual relationship graph based on the scene image features.

[0104] S312: Preprocess the audio modal input to obtain a plurality of mel-spectrogram images, extract a plurality of audio features from the plurality of mel-spectrogram images using a pre-trained audio encoder, and construct an audio relationship graph based on the plurality of audio features.

[0105] In a feasible implementation, in order to better describe the acquisition of personality-related clues from the extracted scene and audio features, the present invention proposes a new method, the structure of which is as follows: Figure 3 As shown, it is used to extract scene-audio correlation features.

[0106] Specifically, the pre-trained CLIP image encoder (ViT) is used to extract scene image features (SI) from scene images. For audio data, the present invention pre-processes the audio data into multiple mel-spectrograms and extracts audio features using a pre-trained VGGish CNN model, resulting in multiple audio features (MA). The present invention then uses these scene image features (SI) to construct a visual relationship graph (VRG).

[0107] In the present invention, each scene image feature is regarded as a node in a complete graph, and the so-called complete graph means that all nodes are connected to each other. Similarly, the present invention can use multiple audio features MA to construct an audio relationship graph ARG. For these two relationship graphs, the present invention uses the cosine similarity between two node features to describe the edge attributes of each relationship graph. Assume that for a relationship graph, the features of any two nodes are , its adjacent nodes are , F is the length of the node feature, then the edge attributes of the two connected nodes are as follows:

[0108] (1)

[0109] S313. Design a graph relationship attention network, obtain visual relationship graph features based on the visual relationship graph and the graph relationship attention network; obtain audio relationship graph features based on the audio relationship graph and the graph relationship attention network.

[0110] In a feasible implementation, in order to explore the association between scene image features and multiple audio features, the present invention designs a graph relation attention network for feature transfer and aggregation of VRG and ARG, such as Figure 4 As shown. It consists of two graph relation attention convolution modules, global maximum pooling, global average pooling, connection operation, one-dimensional convolution module and residual connection. Each graph relation attention convolution module includes a graph attention convolution layer, a RELU activation layer and a one-dimensional convolution module. Among them, the residual connection is used to stabilize the learning process. The global average pooling layer and the global maximum pooling layer are used to independently compress the output features of the second one-dimensional convolution layer into features without time dimension. In this invention, GAT (Graph Attention Network) is used as the basis of the graph attention convolution layer, which is defined as Next, the present invention extracts two relationship graph features based on this, as follows:

[0111] (2)

[0112] (3)

[0113] (4)

[0114] Where, is the input node feature, is the relu activation function, is a one-dimensional CNN convolution function, and are the weight matrix and bias term of CNN. and Represent global maximum pooling and global average pooling respectively, Represents feature concatenation. Finally, the final feature output of the graph relation attention network is:

[0115] (5)

[0116] S314: Concatenate the visual relationship graph features and the audio relationship graph features to obtain scene-audio association features.

[0117] In one feasible implementation, the present invention concatenates the two features described above in the second dimension and fuses them through a one-dimensional convolutional layer. Finally, the present invention concatenates the output features of the visual and audio relationship streams and further integrates the information through a multi-layer perceptron (MLP), consisting of a linear layer, a Relu activation layer, and a dropout layer. Ultimately, the present invention can obtain correlation features between the scene image and audio to represent personality-related information.

[0118] S32: Input the visual modality input into the scene-description word association feature extraction module to obtain the scene-description word association feature.

[0119] In a feasible embodiment, the structure of the CLIPAudio method of the present invention is as follows Figure 5 As shown in Figure 2, it includes two parts: scene-descriptor association feature extraction and audio-descriptor association feature extraction. The CLIPAudio method consists of three parts: 1) personality descriptor list construction; 2) scene-personality descriptor association feature extraction; and 3) audio-personality descriptor association feature extraction.

[0120] Here, we use CLIP's text encoder to generate text description embeddings for these personality descriptors. We also use the "openai / clip-vit-large-patch14-336" model as CLIP's image encoder to generate scene embeddings from the video, and a pre-trained VGG-ish CNN as the audio encoder to generate audio embeddings from the video's audio modality.

[0121] Optionally, the above step S32 may include the following steps S321-S324:

[0122] S321. Construct a personality description list.

[0123] In one possible implementation, the present invention generates personality descriptor embeddings using a list of personality descriptors, which includes 1,608 adjectives that accurately describe human personality traits. This list contains personality terms carefully selected from personality surveys [37,38] and multiple dictionaries [39,40,41] to ensure diversity and representativeness of descriptions, thereby effectively capturing the complexity of human personality.

[0124] S322: Generate personality description word embedding based on the personality description word list and the text encoder.

[0125] S323: Generate scene image embedding based on the visual modality input and the image encoder.

[0126] S324. Calculate the dot product of the personality description word embedding and the scene image embedding to obtain the scene-description word association feature.

[0127] In a feasible implementation, Figure 5 As shown in the upper part of the figure, we propose CLIP-PersD (part of our CLIPAudio method) to extract scene features and personality descriptor features. It consists of three main steps: (1) using CLIP's image encoder to generate embeddings for scene images; (2) using CLIP's text encoder to generate embeddings for personality descriptors; and (3) calculating the similarity score between the scene embedding and the text embedding by calculating their dot product.

[0128] Specifically, first, suppose Indicates inclusion A collection of personality descriptors. Extracted from the video A collection of scene images. For each personality descriptor, there is , this paper uses CLIP's text encoder to generate text embedding ,in At the same time, for each image , the present invention uses CLIP's image encoder to generate image embedding Each image corresponds to a The feature vector of the personality descriptor, where each value represents the similarity score between the image embedding and the corresponding personality descriptor embedding. Here, the present invention calculates their similarity scores by calculating their dot product. Specifically, the present invention obtains a feature matrix , where each entry Indicates the Personality descriptors and The similarity score between images is calculated as the dot product of their corresponding personality descriptors and image embeddings:

[0129] (6)

[0130] Finally, in order to obtain the The present invention needs to average the feature matrices of all images in the video to obtain a single feature vector of the video:

[0131] (7)

[0132] in, It is The feature vector of the video, It is The number of images extracted from a video. In the present invention, the setting .

[0133] S33. Input the audio modal input into the audio-description word association feature extraction module to obtain the audio-description word association feature.

[0134] In a feasible implementation, Figure 5 As shown in the lower part of the paper, this paper proposes a new method named CLIPVGGish for extracting audio description word related features based on CLIP and pre-trained VGGish CNN. The method consists of three main steps: (1) using CLIP's text encoder to generate personality descriptor embeddings; (2) using VGGish CNN encoder to generate audio embeddings of audio cues; (3) calculating the similarity score between personality descriptor embeddings and audio embeddings by calculating the dot product between them.

[0135] Optionally, the above step S33 may include the following steps S331-S333:

[0136] S331. Generate personality description word embedding based on the personality description word list and the text encoder.

[0137] S332: Generate audio embedding based on the audio modality input and the pre-trained audio encoder.

[0138] S333. Calculate the dot product of the personality description word embedding and the audio embedding to obtain the audio-description word association feature.

[0139] In a feasible implementation, the present invention can generate embeddings TP for all personality descriptors P. In addition, the present invention first converts the video into audio data using the ffm-peg tool. After a series of operations such as frame processing, pre-emphasis, window function, fast Fourier transform, Mel filtering, and logarithmic transformation, the audio data is converted into a logarithmic Mel spectrogram. In the present invention, for approximately 15 seconds of video, 25 evenly spaced logarithmic Mel spectrograms M with a shape of [60×64] are extracted. In addition, the VGGish CNN pre-trained on YouTube's AudioSet data is used and fine-tuned as an audio encoder. These Mel spectrograms are input into the pre-trained VGGish CNN model to generate audio embeddings AM.

[0140] Furthermore, each audio embedding corresponds to a feature vector containing N values ​​of personality descriptors, where each value corresponds to an audio embedding The present invention calculates the similarity score between the audio embedding and the corresponding personality description word embedding TP. and the dot product of the description word embedding TP to calculate their similarity scores. Specifically, the present invention can obtain a feature matrix , where each entry Indicates the Personality descriptors and The similarity score between the audio embeddings is calculated as the corresponding personality descriptor And the dot product of the audio embedding AM:

[0141] (8)

[0142] Finally, in order to obtain The present invention averages the feature matrices of all audio embeddings in a video:

[0143] (9)

[0144] Here, It is The feature vector of the video, where |AM| is the The number of audio embeddings extracted from a video. In the present invention, |AM| is set to 25.

[0145] Furthermore, to reduce the dimensionality of the aforementioned scene personality descriptor-related features and audio personality descriptor-related features, the present invention constructed and trained two MLP regression models, each containing two fully connected (FC) layers. Each FC layer consists of a linear layer, a ReLU activation layer, and a dropout layer (parameter p = 0.3), reducing the feature dimensionality from 1608 to 512, and finally to 128. Each linear layer associated with the scene personality descriptor-related features has an additional Bn layer.

[0146] S34. Input the text modal input into the Roberta model to obtain text modal features.

[0147] In one possible implementation, the present invention uses a pretrained Roberta model to extract embedding feature vectors from audio transcripts. This model leverages the parameters of a pretrained Hugging Face model and is fine-tuned on a personality prediction dataset. The following describes the data preprocessing and model architecture of the Roberta model.

[0148] 1) Roberta Preprocessing: Roberta is a variant of the BERT model, using a Transformer architecture with 12 encoder layers and a 12-head self-attention mechanism. Through unsupervised learning, Roberta learns language representations on a much larger dataset than BERT. Notably, Roberta is better able to grasp context and long-range dependencies.

[0149] 2) Model architecture: In the method of the present invention, text processing relies on RobertaTokenizer, which is based on the pre-trained Roberta model. Tokenizer performs several key steps, including: (1) Preprocessing: converting the text to lowercase, removing extra spaces, etc.; (2) Word segmentation: Tokenizer segments the text into tokens based on the constructed word list; (3) Adding special tags: In order to mark the beginning and end of the text, special tags such as [CLS] and [SEP] are added. After these steps, the token sequence is converted into a token index representation. The process of extracting text embedding features using Roberta is as follows: Figure 6 shown.

[0150] S4. Input the scene-audio association features, scene-description word association features, audio-description word association features and text modality features into the feature fusion module to obtain multimodal fusion features.

[0151] like Figure 7 A multimodal attention feature fusion strategy and a regression head based on an MLP model are demonstrated.

[0152] First, a FC layer is used to ensure dimensionality consistency before all features are input into the multimodal attention feature fusion module. For scene description related features, the FC layer structure is slightly different, including a linear layer, a batch normalization layer, a relu activation layer, and a dropout layer with p=0.3. For other input features, the FC layer structure is the same, including a linear layer, a relu activation layer, and a dropout layer with p=0.2.

[0153] Optionally, the above step S4 may include the following steps S41-S43:

[0154] S41. Calculate the scene-audio association features, scene-description word association features, audio-description word association features and text modality features through the first attention score calculation module, the second attention score calculation module, the third attention score calculation module and the fourth attention score calculation module, respectively, to obtain the scene-audio association feature self-attention score, the scene-description word association feature self-attention score, the audio-description word association feature self-attention score and the text modality feature self-attention score.

[0155] S42. Based on the scene-audio associated feature self-attention score, the scene-description word associated feature self-attention score, the audio-description word associated feature self-attention score, and the text modality feature self-attention score, calculate the normalized value of the scene-audio associated feature self-attention score, the normalized value of the scene-description word associated feature self-attention score, the normalized value of the audio-description word associated feature self-attention score, and the normalized value of the text modality feature self-attention score.

[0156] S43. Calculate the multimodal fusion feature based on the normalized value of the scene-audio associated feature self-attention score, the normalized value of the scene-description word associated feature self-attention score, the normalized value of the audio-description word associated feature self-attention score, and the normalized value of the text modality feature self-attention score.

[0157] In a feasible implementation, the present invention designs a multimodal attention feature fusion strategy to fuse four features. First, four attention score calculation modules with the same structure (parameters are not shared) are used to obtain the self-attention scores of four different features. Then, the normalized values ​​of the attention scores of the four different features are obtained by dividing each self-attention score by the sum of the self-attention scores of all modalities. Finally, the attention scores are multiplied by the feature vectors of different features and summed to form the final multimodal fusion feature. Through the proposed strategy, the present invention aims to capture the importance based on the complex relationship between different modal features. The feature attention module includes a fully connected layer with a linear layer and a tanh function, and another layer with a linear layer and a sigmoid function. Assume that the obtained specific modality representation vector is ,but The self-attention score is recorded as For the four features, their self-attention scores can be calculated as follows:

[0158] (10)

[0159] in, and Represent the learnable parameters of the two fully connected layers (parameters between different modalities are not shared). The tanh and sigmoid functions are used to limit the attention weights to the interval [-1, 1]. Then, the obtained attention vector (* represents any one of {T, VT, AT, VA}) and performs element-wise multiplication with each dimension of the multimodal fusion feature head h. The formula is as follows:

[0160] (11)

[0161] (12)

[0162] (13)

[0163] Finally, in order to obtain more robust multimodal fusion features, the present invention uses a multi-head mechanism to generate multiple fusion features and connects them to form the final multimodal fusion feature :

[0164] (14)

[0165] S5. Input the multimodal fusion features into the perception prediction module to obtain the personality perception results.

[0166] In this paper, a two-layer perceptron (MLP) regression model is used as a personality perception model. This perception model includes a FC module and an MLP module. The FC module consists of a linear layer, a Relu activation layer, and a dropout layer to reduce the dimensionality of the multimodal fusion feature vector. The multimodal fusion feature vector is then processed by a single-layer MLP model to predict scores for different personality traits, with output values ​​between 0 and 1. The MLP model consists of a linear layer and a sigmoid activation layer.

[0167] Furthermore, in order to optimize the overall framework, the present invention uses the Bell loss function, which is defined as follows:

[0168] (15)

[0169] in, and are the true labels and predicted values, is the derivative parameter that controls the magnitude of the change. resulting in higher gradients. Is a scale parameter that controls the range of the loss function. In the present invention, , .

[0170] Using the multimodal personality perception method based on multiple correlation features and graph relational attention networks disclosed in the present invention, the system can intelligently and quickly predict people's Big Five personality traits. This method can be widely used in job interviews, purchasing behavior, national health (the pathogenesis and rehabilitation of mental illnesses such as depression and anxiety), education, talent recruitment, crime prevention, and other aspects.

[0171] Automatic personality trait perception leverages multimodal information to accurately predict personality traits, such as behavior and appearance, from videos. Existing large-scale multimodal pre-trained models excel in downstream tasks. However, existing methods overlook the intermodal correlation features and the unique capabilities of graph-relational attention networks in extracting critical temporal information. Therefore, this paper proposes a novel multimodal framework for personality assessment based on multiple correlation features and a graph-relational attention network. This paper also proposes a method for extracting scene-audio correlation features through graph-relational representation learning. Furthermore, this paper designs a CLIPAudio method that extends CLIP to extract correlation features between scene personality descriptors and audio personality descriptors. Specifically, this paper explores the correlation between personality descriptors and visual and audio modalities using cross-modal and intramodal similarity. Furthermore, this paper utilizes a pre-trained RoBERTa model as the backbone of a text-modality-specific network to extract textual cues related to personality traits. Finally, this paper designs a multimodal attention feature fusion strategy to combine these features and train a regression model to predict personality traits.

[0172] In an embodiment of the present invention, a multimodal attention fusion framework is proposed for personality perception. The framework utilizes multiple correlation features and graph relation attention networks and is designed specifically for vision-language-audio processing. The present invention includes audio modalities in a unified manner by extending the vision-language pre-training model CLIP. The present invention utilizes contrastive learning techniques from the perspectives of cross-modality and intra-modality to fully utilize the multimodal pre-training model. In order to extract scene-audio correlation features, a scene-audio correlation feature extraction method is proposed in the framework of the present invention. In addition, the present invention extends the CLIP model to CLIPAudio audio, which is used to mine the correlation features of scene personality descriptors and audio personality descriptors respectively. The present invention also utilizes the pre-trained Roberta model to extract text-specific features. In order to effectively fuse these multimodal information, the present invention designs a multimodal attention feature fusion strategy.

[0173] Figure 8 This is a block diagram of a multimodal personality perception device based on multiple correlation features and graph relationship attention according to an exemplary embodiment. The device is used in a multimodal personality perception method based on multiple correlation features and graph relationship attention. Figure 8 The device includes an acquisition module 310, a data preprocessing module 320, a modal feature extraction network module 330, a feature fusion module 340 and an output module 350.

[0174] The acquisition module 310 is used to acquire an input video for personality perception.

[0175] The data preprocessing module 320 is used to input the input video into the data preprocessing module to obtain visual modality input, audio modality input and text modality input.

[0176] The modal feature extraction network module 330 is used to input visual modal input, audio modal input and text modal input into the modal feature extraction network module to obtain scene-audio association features, scene-description word association features, audio-description word association features and text modal features.

[0177] The feature fusion module 340 is used to input the scene-audio association feature, the scene-description word association feature, the audio-description word association feature and the text modality feature into the feature fusion module to obtain a multimodal fusion feature.

[0178] The output module 350 is used to input the multimodal fusion features into the perception prediction module to obtain the personality perception result.

[0179] In an embodiment of the present invention, a multimodal attention fusion framework is proposed for personality perception. The framework utilizes multiple correlation features and graph relation attention networks and is designed specifically for vision-language-audio processing. The present invention includes audio modalities in a unified manner by extending the vision-language pre-training model CLIP. The present invention utilizes contrastive learning techniques from the perspectives of cross-modality and intra-modality to fully utilize the multimodal pre-training model. In order to extract scene-audio correlation features, a scene-audio correlation feature extraction method is proposed in the framework of the present invention. In addition, the present invention extends the CLIP model to CLIPAudio audio, which is used to mine the correlation features of scene personality descriptors and audio personality descriptors respectively. The present invention also utilizes the pre-trained Roberta model to extract text-specific features. In order to effectively fuse these multimodal information, the present invention designs a multimodal attention feature fusion strategy.

[0180] Figure 9 is a structural diagram of a multimodal personality perception device provided by an embodiment of the present invention, such as Figure 9 As shown, the multimodal personality perception device may include the above Figure 8 Optionally, the multimodal personality perception device 410 may include a first processor 2001.

[0181] Optionally, the multimodal personality perception device 410 may further include a memory 2002 and a transceiver 2003 .

[0182] The first processor 2001, the memory 2002 and the transceiver 2003 may be connected via a communication bus.

[0183] The following combination Figure 9The components of the multimodal personality perception device 410 are described in detail:

[0184] The first processor 2001 is the control center of the multimodal personality perception device 410 and can be a single processor or a collective term for multiple processing elements. For example, the first processor 2001 can be one or more central processing units (CPUs), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention, such as one or more digital signal processors (DSPs) or one or more field programmable gate arrays (FPGAs).

[0185] Optionally, the first processor 2001 may execute various functions of the multimodal personality perception device 410 by running or executing a software program stored in the memory 2002 and calling data stored in the memory 2002 .

[0186] In a specific implementation, as an embodiment, the first processor 2001 may include one or more CPUs, such as Figure 9 CPU0 and CPU1 are shown in FIG.

[0187] In a specific implementation, as an embodiment, the multimodal personality perception device 410 may also include multiple processors, such as Figure 9 1 and 2. The first processor 2001 and the second processor 2004 are shown in FIG. Each of these processors can be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). A processor herein can refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0188] The memory 2002 is used to store the software program for executing the solution of the present invention, and is controlled by the first processor 2001 for execution. The specific implementation method can refer to the above method embodiment and will not be repeated here.

[0189] Alternatively, the memory 2002 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, a random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, an optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and capable of being accessed by a computer, but is not limited thereto. The memory 2002 may be integrated with the first processor 2001 or may exist independently and be accessed through the interface circuit ( Figure 9 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0190] The transceiver 2003 is used to communicate with a network device or a terminal device.

[0191] Optionally, the transceiver 2003 may include a receiver and a transmitter ( Figure 9 The receiver is used to implement a receiving function, and the transmitter is used to implement a sending function.

[0192] Optionally, the transceiver 2003 may be integrated with the first processor 2001, or may exist independently and communicate with the first processor 2001 through the interface circuit ( Figure 9 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0193] It should be noted that Figure 9 The structure of the multimodal personality perception device 410 shown in the figure does not constitute a limitation on the router. The actual knowledge structure recognition device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0194] In addition, the technical effects of the multimodal personality perception device 410 can refer to the technical effects of the multimodal personality perception method based on multiple correlation features and graph relationship attention described in the above method embodiment, and will not be repeated here.

[0195] It should be understood that the first processor 2001 in the embodiment of the present invention may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor, or the processor may be any conventional processor, etc.

[0196] It should also be understood that the memory in the embodiments of the present invention may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory may be random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0197] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A multimodal personality perception method based on multiple correlation features and graph relationship attention, characterized by: The method comprises: S1. Obtaining an input video for personality perception; S2. Inputting the input video into a data preprocessing module to obtain visual modality input, audio modality input, and text modality input; S3, inputting the visual modality input, the audio modality input, and the text modality input into a modality feature extraction network module to obtain scene-audio association features, scene-description word association features, audio-description word association features, and text modality features; S4, inputting the scene-audio association feature, the scene-description word association feature, the audio-description word association feature, and the text modality feature into a feature fusion module to obtain a multimodal fusion feature; S5. Inputting the multimodal fusion features into a perception prediction module to obtain a personality perception result; The step S3 of inputting the visual modality input, the audio modality input, and the text modality input into a modality feature extraction network module to obtain scene-audio association features, scene-description word association features, audio-description word association features, and text modality features includes: S31, inputting the visual modality input and the audio modality input into a scene-audio correlation feature extraction module to obtain scene-audio correlation features; S32, inputting the visual modality into a scene-description word association feature extraction module to obtain scene-description word association features; S33, inputting the audio modality into an audio-description word association feature extraction module to obtain an audio-description word association feature; S34, inputting the text modality into the Roberta model to obtain text modality features; The step S31 of inputting the visual modality input and the audio modality input into a scene-audio correlation feature extraction module to obtain scene-audio correlation features includes: S311. Extracting scene image features from the visual modality input using a pre-trained image encoder, and constructing a visual relationship graph based on the scene image features; S312: Preprocess the audio modality input to obtain a plurality of mel-spectrogram images, extract a plurality of audio features from the plurality of mel-spectrogram images using a pre-trained audio encoder, and construct an audio relationship graph based on the plurality of audio features; S313. Designing a graph relation attention network, obtaining visual relation graph features based on the visual relation graph and the graph relation attention network; obtaining audio relation graph features based on the audio relation graph and the graph relation attention network; S314: Concatenate the visual relationship graph features and the audio relationship graph features to obtain scene-audio correlation features; The step of inputting the visual modality into a scene-description word association feature extraction module in step S32 to obtain scene-description word association features includes: S321, construct a list of personality descriptors; S322, generating a personality description word embedding according to the personality description word list and the text encoder; S323, generating a scene image embedding according to the visual modality input and the image encoder; S324, calculating the dot product of the personality description word embedding and the scene image embedding, thereby obtaining a scene-description word association feature; The step S33 of inputting the audio modality into an audio-description word association feature extraction module to obtain audio-description word association features includes: S331, generating personality description word embeddings based on the personality description word list and the text encoder; S332: Generate an audio embedding according to the audio modality input and a pre-trained audio encoder; S333. Calculate the dot product of the personality description word embedding and the audio embedding to obtain audio-description word association features.

2. The multimodal personality perception method based on multiple correlation features and graph relationship attention according to claim 1 is characterized in that: The step S4 of inputting the scene-audio association feature, the scene-description word association feature, the audio-description word association feature, and the text modality feature into a feature fusion module to obtain a multimodal fusion feature includes: S41, calculating the scene-audio association feature, the scene-description word association feature, the audio-description word association feature, and the text modality feature through a first attention score calculation module, a second attention score calculation module, a third attention score calculation module, and a fourth attention score calculation module, respectively, to obtain a scene-audio association feature self-attention score, a scene-description word association feature self-attention score, an audio-description word association feature self-attention score, and a text modality feature self-attention score; S42, calculating, based on the scene-audio association feature self-attention score, the scene-description word association feature self-attention score, the audio-description word association feature self-attention score, and the text modality feature self-attention score, a normalized value of the scene-audio association feature self-attention score, a normalized value of the scene-description word association feature self-attention score, a normalized value of the audio-description word association feature self-attention score, and a normalized value of the text modality feature self-attention score; S43. Calculate the multimodal fusion feature based on the normalized value of the scene-audio association feature self-attention score, the normalized value of the scene-description word association feature self-attention score, the normalized value of the audio-description word association feature self-attention score, and the normalized value of the text modality feature self-attention score.

3. A multimodal personality perception device based on multiple correlation features and graph relationship attention, wherein the multimodal personality perception device based on multiple correlation features and graph relationship attention is used to implement the multimodal personality perception method based on multiple correlation features and graph relationship attention as claimed in any one of claims 1-2, characterized in that: The device comprises: An acquisition module, used to acquire an input video to be subjected to personality perception; A data preprocessing module, configured to input the input video into the data preprocessing module to obtain visual modality input, audio modality input, and text modality input; A modal feature extraction network module is used to input the visual modal input, the audio modal input, and the text modal input into the modal feature extraction network module to obtain scene-audio association features, scene-description word association features, audio-description word association features, and text modal features; A feature fusion module, configured to input the scene-audio association feature, the scene-description word association feature, the audio-description word association feature, and the text modality feature into the feature fusion module to obtain a multimodal fusion feature; The output module is used to input the multimodal fusion features into the perception prediction module to obtain personality perception results.

4. A multimodal personality perception device, characterized in that: The multimodal personality perception device comprises: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the method according to any one of claims 1 to 2 is implemented.

5. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program code, which can be called by a processor to execute the method according to any one of claims 1 to 2.

Images