Processing method and electronic equipment

Abstract

The invention provides a processing method. The processing method comprises the following steps: acquiring target data of a plurality of objects in a conference event; based on the target data, relation information among the multiple objects is determined, and the relation information represents an emotion relation or an action relation between any object in the multiple objects and other objects; target information of the conference event is generated based on the relation information among the multiple objects, and the target information is used for displaying the interaction state in the conference event.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and more specifically, to a processing method and an electronic device. Background Technology

[0002] With the rapid development of intelligent meeting systems and multimedia analytics, automated understanding and analysis of meeting scenarios has become an important direction for improving collaboration efficiency. However, current analysis of meeting scenarios mostly focuses on individual participants, failing to reflect the true overall atmosphere and deep interpersonal relationships within the meeting, thus limiting the application of meeting analytics in decision support and situational awareness. Summary of the Invention

[0003] In view of this, the present disclosure provides a processing method and an electronic device.

[0004] One aspect of this disclosure provides a processing method, comprising: acquiring target data of multiple objects in a meeting event; determining relationship information between the multiple objects based on the target data, the relationship information representing the emotional or actional relationship between any one of the multiple objects and other objects; and generating target information of the meeting event based on the relationship information between the multiple objects, the target information being used to display the interactive state in the meeting event.

[0005] Another aspect of this disclosure provides a processing apparatus, comprising: a first acquisition module for acquiring target data of a meeting event, the target data indicating multiple objects of the meeting event; a first determination module for determining relational information of the objects based on the target data, the relational information representing the emotional or actional relationship of the objects with respect to at least one of the multiple objects; and a first generation module for generating target information of the meeting event based on the objects and the relational information of the objects, the target information indicating the interactive state in the meeting event.

[0006] Another aspect of this disclosure provides an electronic device, including: at least one processor; and a memory connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform at least one of the following operations: acquiring target data of multiple objects in a meeting event; determining relationship information between the multiple objects based on the target data, the relationship information representing the emotional or actional relationship between any one of the multiple objects and other objects; and generating target information for the meeting event based on the relationship information between the multiple objects, the target information being used to display the interactive state in the meeting event.

[0007] Another aspect of this disclosure provides a computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to perform a processing method according to any of the foregoing embodiments.

[0008] Another aspect of this disclosure provides a computer program product, including a computer program / instructions, characterized in that the computer program / instructions, when executed by a processor, implement the operation of the processing method of any of the foregoing embodiments. Attached Figure Description

[0009] The above and other objects, features and advantages of this disclosure will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0010] Figure 1 A flowchart illustrating a processing method according to an embodiment of the present disclosure is shown schematically.

[0011] Figure 2A A flowchart illustrating the generation of target information for a conference event in a processing method according to an embodiment of the present disclosure is shown schematically.

[0012] Figure 2B A target atlas is schematically illustrated according to an embodiment of the present disclosure;

[0013] Figure 3 This schematically illustrates another flowchart of the process for generating target information of a conference event in an embodiment of the present disclosure;

[0014] Figure 4 A schematic diagram illustrating a hybrid meeting scenario according to an embodiment of the present disclosure is shown.

[0015] Figure 5 Another flowchart illustrating a processing method according to an embodiment of the present disclosure is shown schematically;

[0016] Figure 6 Another flowchart illustrating a processing method according to an embodiment of the present disclosure is shown schematically;

[0017] Figure 7 This illustration schematically shows the presentation of target image information in an interactive window according to an embodiment of the present disclosure;

[0018] Figure 8 This schematically illustrates an interface diagram showing the presentation of emotional information of a single object in an interactive window according to an embodiment of the present disclosure;

[0019] Figure 9 Another flowchart illustrating a processing method according to an embodiment of the present disclosure is shown schematically;

[0020] Figure 10 A block diagram schematically illustrates a processing apparatus according to embodiments of the present disclosure; and

[0021] Figure 11A block diagram of an electronic device suitable for implementing the methods described above, according to embodiments of the present disclosure, is illustrated schematically. Detailed Implementation

[0022] The embodiments of the present disclosure will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the disclosure. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.

[0023] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0024] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0025] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0026] In the embodiments disclosed herein, the collection, updating, analysis, processing, use, transmission, provision, disclosure, and storage of data (e.g., including but not limited to user personal information) comply with relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. In particular, necessary measures have been taken to prevent unauthorized access to user personal information data and to safeguard user personal information security, network security, and national security.

[0027] Embodiments of this disclosure provide a processing method, including: acquiring target data of multiple objects in a meeting event; determining relationship information between the multiple objects based on the target data, wherein the relationship information characterizes the emotional or actional relationship between any one of the multiple objects and other objects; and generating target information of the meeting event based on the relationship information between the multiple objects, wherein the target information is used to display the interactive state in the meeting event.

[0028] Figure 1 A flowchart illustrating a processing method according to an embodiment of the present disclosure is shown schematically.

[0029] like Figure 1 As shown, the processing method may include at least operations S110 to S130.

[0030] In operation S110, target data of multiple objects in a meeting event is acquired. A meeting event can be any form of multi-party information interaction scenario, including but not limited to: offline physical meetings based on physical space, remote video / telephone conferences based on network communication technology, virtual reality conferences that integrate virtual reality or augmented reality technology, or hybrid conferences of the above.

[0031] Multiple objects typically refer to the main entities participating in the meeting event. In most cases, objects refer to the attendees (users). In some scenarios, objects may also include, but are not limited to, interactive virtual digital humans, intelligent agents, or meeting devices endowed with anthropomorphic attributes within the meeting system.

[0032] Target data refers to the set of basic data that reflects the presence, behavior, and environmental context of these objects in the meeting event.

[0033] Target data can be acquired through real-time recording using various sensors and data acquisition devices deployed in the meeting environment, or by receiving data streams uploaded from remote terminals via network interfaces. For example, this could include video stream data containing all participants, captured by a panoramic camera installed in the center of the meeting room; audio data containing the speeches of each participant, captured by a microphone array; or participant login information and text or image chat logs obtained from the meeting software backend.

[0034] In operation S120, based on target data, relationship information between multiple objects is determined. This relationship information characterizes the emotional or behavioral relationships between any one object and other objects. Relationship information refers to the direct or indirect connections between different objects within the specific time and space of a meeting. This connection differs from the independent attributes of an object itself (e.g., someone is happy), but rather focuses on the directed link between objects (e.g., someone is happy for someone else). Relationship information can focus on physical or behavioral behavioral relationships, reflecting an object's current focus of attention, the direction of communication channels, or the goal of information transmission. Relationship information can also be emotional relationships at the logical, psychological, or emotional levels, reflecting an object's attitude, reaction, emotional polarity, or psychological tendency towards another object.

[0035] By extracting feature information from target data and analyzing behavioral patterns, communication content, relative positions, or logical instructions between objects, it is possible to determine whether there is a connection from a source object to one or more target objects, thereby identifying such paired relationship information. For example, based on the behavioral characteristics of the first object in a meeting, its current focus can be analyzed to determine whether the first object has a focus or referential action relationship with the second object; or, based on the feedback signals or expressions of the first object to the second object, its emotional tendency can be analyzed to determine whether the first object holds a specific polarity (such as positive, negative, or neutral) emotional relationship with the second object.

[0036] In operation S130, target information for the meeting event is generated based on the relationship information between multiple objects. This target information is used to display the interactive state within the meeting event. Target information is the result of comprehensively processing, abstracting, or reorganizing the complex many-to-many relationships within the meeting event. Target information is no longer fragmented data points, but rather high-level information that reflects the overall situation or local characteristics of the meeting.

[0037] The interactive state in a meeting event refers to the social network state, interactive atmosphere, or dynamic characteristics constituted by information about multiple objects and their relationships with each other. It reflects the activity level of meeting interactions, the main flow of influence, the emotional tone of the group, or the current communication pattern.

[0038] Generating target information can involve aggregating and statistically analyzing multiple defined relationships, performing structured modeling, weighted calculations, or mapping states to obtain the final analysis results. The final form can be a data structure that can be further recognized by a computer system, or a signal that can be intuitively understood by the user. For example, generating structured data to characterize the distribution of interactions between multiple objects; generating numerical indicators to quantify the intensity or closeness of interactions between objects; or generating identifying information to qualitatively describe the type of interaction or social atmosphere characteristics of the current meeting.

[0039] According to embodiments of this disclosure, by acquiring target data and deeply mining the implicit inter-object relationship information, the limitations of isolated feature recognition for a single object (such as simply recognizing someone's facial expression or simply recording someone's speech) are overcome. An "object-to-object" relational dimension is introduced, enabling the capture of subtle interaction flows, attention flows, and emotional projections between people in a meeting. By transforming these discrete individual data into target information reflecting the overall interaction state, the intangible social context in the physical or digital world can be effectively digitally revealed, thereby improving meeting decision-making efficiency and enhancing social perception capabilities.

[0040] Figure 2AA flowchart illustrating the generation of target information for a conference event in a processing method according to an embodiment of the present disclosure is shown. Figure 2B The target map is illustrated schematically according to an embodiment of the present disclosure.

[0041] like Figure 2A As shown, based on the aforementioned embodiments, operation S130 may include operation S210.

[0042] In operation S210, a target graph is generated based on the relationship information between multiple objects; the target information of the conference event includes the target graph. The target graph includes: nodes, used to represent objects; and edges, used to represent the relationship information between multiple objects. A target graph is a data structure model used to describe entities and their interrelationships, typically consisting of a set of nodes and a set of edges connecting these vertices. In this embodiment, the target graph can be an image for display, or it can be a computer-readable, storable, and dynamically updated topological structure data; it is also a mathematical structure that mathematically represents the social network in the conference scenario.

[0043] like Figure 2B As shown, there are multiple objects in the meeting event, which are mapped to nodes A, B, C, D, E, F, and G in the graph. Simultaneously, based on the established relationships between the objects, connections, i.e., edges, are established between the corresponding nodes. If it is determined that objects B, C, and E all have relationships with object A, then a directed edge is generated from node B to node A, a directed edge is generated from node C to node A, and a directed edge is generated from node E to node A. If it is determined that object D has a relationship with object B, then a directed edge is generated from node D to node B. If it is determined that there is a bidirectional relationship between object A and object F, then a bidirectional edge (or two unidirectional edges in opposite directions) connecting node A and node F is generated. If it is determined that object G currently has no relationships with any other objects, then node G is represented as an isolated node in the graph.

[0044] According to embodiments of this disclosure, by constructing a target graph, complex meeting relationship information is transformed into an intuitive and structured network model, which systematically integrates discrete objects and relationships in the meeting. By generating target information describing the graph structure, standard underlying data is provided for subsequent data transmission, storage, further processing, or visualization, enabling the social structure of the meeting to be recorded and reproduced in a digital form.

[0045] In addition, as a parallel implementation of generating the above-mentioned topology data, the target information can also be represented in other data forms that can reflect the interaction state, such as, but not limited to, data matrices, text summaries, and state identifiers.

[0046] For example, a two-dimensional array or matrix (such as an adjacency matrix) can be generated containing the interactions between objects at the current moment. In this matrix, rows and columns correspond to the objects in the meeting, and the values ​​in the matrix cells indicate whether there is a relationship between the corresponding two objects. For example, if the first object and the second object interact, the corresponding matrix element is set to a specific value (such as 1); if there is no interaction, it is set to 0.

[0047] For example, natural language generation technology can be used to output a log text describing the current flow of interactions in a meeting. This text directly records the directional relationships between objects. For instance, the generated text might read, "Object A is following Object B; Object C is interacting bidirectionally with Object D; Object E is in a non-interactive state."

[0048] For example, a feature code or status label indicating the current overall interaction pattern can be generated based on the overall distribution of relationship information. For instance, a specific code can be generated that corresponds to a predefined social network pattern such as "many-to-one attention pattern," "pair-to-pair group discussion pattern," or "all-member discrete pattern."

[0049] Figure 3 Another flowchart illustrating the process of generating target information for a conference event in a processing method according to an embodiment of the present disclosure is shown.

[0050] The target data includes at least one of the following: text data, audio data, image data, video data, and interaction data related to the meeting event. Image and video data typically contain the visual appearance and motion frames of objects; audio data contains speech content and intonation characteristics; text data contains transcribed text or system logs; and interaction data contains records of objects' actions on the meeting software.

[0051] like Figure 3 As shown, based on the aforementioned embodiments, operation S210 may include operations S310 to S340.

[0052] In operation S310, multiple objects are identified as nodes in the target graph. First, it's necessary to identify the valid participants from the target data and assign a unique identifier to each object. Different objects can be distinguished using face detection, voiceprint recognition, or login ID verification, and corresponding node instances can be created in the graph database. For example, a face recognition algorithm can be used to detect participants A, B, and C from video frames and use them as corresponding nodes in the graph.

[0053] In operation S320, based on target data, emotional and / or action information of multiple objects is determined. For emotional information, image data can be analyzed using a facial expression recognition model, or audio data can be analyzed using a speech emotion analysis model to obtain classification labels such as "happy," "angry," or "neutral," or corresponding confidence scores. For action information, facial and limb feature points of the object can be located using a keypoint detection algorithm to calculate the object's posture, or the object's posture can be obtained by analyzing the target data using a large visual analysis model. For example, for object A, its facial expression is identified as "focused" (emotional information), and its head is detected to be turned 30 degrees to the right and its eyes are moving to the right (action information).

[0054] In operation S330, based on the emotional and / or action information of the first object among multiple objects, the first edge of the first node corresponding to the first object is determined. The edge is used to indicate the relationship information between the first object and other objects. Based on the action direction or emotional projection exhibited by the first object, it is determined whether it has established a connection with other objects in the field, and this connection is mapped to an edge in the graph. For example, if the action of object A indicates that it is interacting with object B, then an edge connecting node A and node B is created in the graph.

[0055] In operation S340, a target graph is generated based on at least one first node and at least one first edge. All node instances and all generated edge instances determined in the preceding steps are combined to construct a complete graph data object.

[0056] Building upon the foregoing embodiments, the motion information includes at least one of the following: the object's gaze, head posture, body posture, and actions. Gaze and head posture can be used to indicate the object's current, subtle visual focus, such as precisely locking onto a speaker or presentation screen through eye movement and facial orientation. Body posture (such as torso orientation and shoulder angle) can be used to indicate the object's long-term, macro-level social tendencies; for example, facing someone directly usually indicates an open and receptive willingness to communicate, while turning to the side or away indicates avoidance or rejection. Actions (such as pointing or waving) can be used to indicate specific, explicit interactive intentions; for example, pointing at someone explicitly directs the attention of the entire room to the pointed-out object.

[0057] Operation S330 may include: determining the attention vector of a first object based on action information; in response to the attention vector of the first object pointing to a second object among multiple objects, determining the first edge of the first node corresponding to the first object pointing to the second node corresponding to the second object; and determining the attributes of the first node corresponding to the first object based on emotion information.

[0058] Based on motion information, the attention vector of the first object is determined. The attention vector is a virtual ray that integrates head posture, gaze direction, body orientation, or limb movements, used to simulate the object's perceptual projection path or intention indication path in three-dimensional space. Typically, the ray is constructed by weighted fusion of head posture vectors and eye gaze vectors. In other scenarios, a person's attention direction may align with the direction of their ears; for example, when someone is attentively listening to another, they may turn their ear towards the target rather than focusing their gaze. In this case, the auditory projection path of the object in three-dimensional space is simulated based on the ear's outline and orientation as the attention vector. Furthermore, when motion information includes body posture, the attention vector can be determined based on the orientation of the torso. For example, when the face cannot be detected (e.g., facing away from the camera), or to determine the object's overall orientation, the attention vector is determined by the direction of the perpendicular bisector of the line connecting the detected shoulder keypoints, or the direction of the normal to the chest plane, starting from the center of the body and extending forward. When motion information includes specific actions (e.g., indicative gestures), the attention vector can be constructed based on keypoints of the limb skeleton. For example, when the first object is identified as making a "pointing" gesture, a ray extending from the fingertip is fitted based on the key point coordinates of the elbow, wrist, and fingertip.

[0059] In response to the attention vector of the first object pointing to the second object among multiple objects, the first edge pointing from the first node corresponding to the first object to the second node corresponding to the second object is determined. It then checks whether the attention vector (ray) intersects or collides with the regions occupied by other objects in the meeting space. If the ray passes through the region occupied by the second object, or if the angle between the ray and the position of the second object is less than a preset threshold, it is determined that the first object is "paying attention" to the second object, thus generating a directed edge from the first node to the second node in the graph. For example, if the attention vector of object A passes through the head region of object B, then an edge from A to B is generated.

[0060] Based on emotional information, the attributes of the first node corresponding to the first object are determined. The identified emotional labels or values ​​are mapped to the attribute values ​​of nodes in the graph. For example, if the emotional identification result of object A is "anger", then the attributes of node A are set to {emotion: "negative", type: "anger", intensity: 0.8}.

[0061] Operation S340 may include: generating a target graph based on at least one first node, a second node, and at least one first edge, and the attributes of the first node. A complete graph data is constructed by combining multiple nodes and edges obtained from the preceding steps.

[0062] According to embodiments of this disclosure, by fusing multi-dimensional action and emotion information to obtain an attention vector, it is possible to accurately capture implicit interaction flows such as gaze, hearing, and gestures. Simultaneously, by assigning emotion information to node attributes, the graph can reflect the emotional nuances behind the interaction. This combination not only identifies "who is paying attention to whom" but also further infers whether this attention is "appreciative listening" or "hostile confrontation," thereby significantly improving the accuracy and depth of meeting analysis.

[0063] Figure 4 A schematic diagram of a hybrid meeting scenario according to an embodiment of the present disclosure is shown.

[0064] Based on the aforementioned embodiments, the meeting event includes a first meeting scenario and a second meeting scenario. The second meeting scenario is presented within the first meeting scenario through an interactive window. The target data also includes interaction information between the object and the interactive window. The first meeting scenario typically refers to the current local physical space, while the second meeting scenario typically refers to a virtual or physical meeting space accessed remotely. The interactive window is the medium for presenting the remote scenario to the local device, typically manifested as a display screen, a projected image, or a holographic projection area. Interaction information refers to the proactive information executed by the local object in response to the interactive window, used to assist in determining explicit relationships between objects. Examples include gaze information, operation information executed in response to the interactive window, such as operation information corresponding to substantive information transmission behaviors initiated to remote objects via keyboard, mouse, touchscreen, microphone, etc.

[0065] like Figure 4 As shown, the first meeting scenario is a local meeting room, including objects A, B, C, and D; the second meeting scenario is a remote access point, including objects E, F, and G. The interaction window is a large display screen mounted on the wall, displaying the video feeds of objects E, F, and G respectively. Interaction information may include object B looking at object F's avatar on the screen, text messages sent by object C to object G, image files, or directed audio streams, etc.

[0066] In response to the attention vector of the first object pointing to the second object among the plurality of objects, determining that the first node corresponding to the first object points to the first side of the second node corresponding to the second object may further include: in response to the attention vector of the first object pointing to the second object in the interaction window, determining that the first node corresponding to the first object points to the first side of the second node corresponding to the second object based on the attention vector and / or the interaction information.

[0067] When the attention of the first object (local user) is directed to the interactive window (screen), it is necessary to analyze which second object (remote user) in the window it is specifically focusing on.

[0068] In one implementation, edges can be determined based on attention vectors, identifying the intersection point between the attention vector of the first object (e.g., a gaze ray) and the plane containing the interactive window in three-dimensional space. Subsequently, the current interface layout information of the interactive window is obtained, defining the display area coordinates of each object within the window. If the intersection point falls within the area displaying the second object, it is determined that the attention vector of the first object points to the second object. Alternatively, computer vision technology can be used to identify the face or body shape of the second object displayed in the interactive window as a "virtual entity" in the environment, and to detect whether the attention vector directly points to this virtual entity.

[0069] In one implementation, interaction information determines edges. Interaction information refers to the record of actions performed by a first object on elements within an interaction window via a human-computer interaction device (such as a mouse, touchscreen, or VR controller). When attention vectors (such as gaze) are difficult to capture precisely or are obstructed, interaction information serves as a direct basis for determining attention relationships. It monitors whether the first object performs operations such as selection, zooming, pinning, or data transmission on a specific area within the interaction window. If the operation corresponds to a second object, then the first edge pointing from the first node to the second node is determined. It should be noted that those skilled in the art can also combine these two or other methods to determine edges.

[0070] Furthermore, a two-way data synchronization mechanism can be used to determine how a remote second object (e.g., E, F, G) follows a local first object (e.g., A, B, C, D). Just as the local system calculates the gaze of a local user, the remote system also calculates the gaze direction of the second object within its environment. When the remote system determines that the second object's gaze is directed at the "avatar of the local object on the remote screen," it generates a corresponding attention command and sends it to the local system. Upon receiving this command, the local system can then construct an edge from the second object to the first object.

[0071] For example, combining Figure 4 For "local viewing of remote": If it is detected that object A's gaze is projected onto the area of ​​object E displayed on the screen (or the 360-degree camera directly recognizes that A is looking at E on the screen), then a directed edge from node A to node E is generated in the graph. For "interactive information assistance": If it is detected that object B sends an image to object F, then an edge from node B to node F is directly generated based on this interactive information. For "remote viewing of local": If the remote end detects that object G turns its head to look at the window displaying object C on the screen, the remote end sends this "data of interest" back to the local system, and the local system generates a directed edge from node G to node C accordingly.

[0072] By establishing a spatial mapping relationship between attention vectors and the display area of ​​interactive windows, or by recognizing screen content as environmental entities through visual technology, eye tracking in physical space is effectively extended to digital screen space. Combined with explicit interaction information and a remote data synchronization mechanism, this overcomes the limitation of traditional eye-tracking analysis, which can only handle local entity interactions, and can accurately capture cross-spatial connections between local and remote objects. This ensures that the generated map fully covers all participants in a hybrid meeting scenario, achieving seamless perception and unified modeling of online and offline integrated social networks.

[0073] Based on the foregoing embodiments, the processing method may further include: determining the emotional information of the object based on the target data, and operation S210 may include: generating a target map based on multiple objects and their relationship information and emotional information.

[0074] Based on the target data, determine the emotional information of the participants. Emotional information refers to characteristic data reflecting the participants' current psychological state, emotional polarity, or cognitive response during the meeting. It is a qualitative or quantitative description of how the participants "participated in the meeting." Emotional information can be determined using multimodal sentiment analysis techniques. For example, analyzing facial micro-expressions based on image data; analyzing speech tone, loudness, or speech rate based on audio data; or analyzing the semantic sentiment tendency of the participants' speech content based on text data. Alternatively, pre-trained sentiment analysis models can be used to process the target data and obtain the emotional information of each participant. Emotional information includes, but is not limited to, happiness, contemplation, aversion, anger, and calmness.

[0075] A target graph is generated based on information about multiple objects, their relationships, and their emotional states. The attributes of nodes represent the emotional information of the objects. The generated target graph is further defined as an attribute graph, where nodes are not merely topological connection points but also containers carrying metadata. It maps specific emotional information to the attribute fields of corresponding nodes, enabling the graph to simultaneously express social structure and individual states. Specifically, emotional information can be stored as node text labels, enumerated status codes, or quantified emotional value vectors. For example, in the generated target graph, the attribute field of node A is labeled {Emotion: "Excited"}, the attribute field of node B is labeled {Emotion: "Bored"}, and the attribute field of node C is labeled {Emotion: "Neutral"}.

[0076] According to embodiments of this disclosure, by identifying the emotional information of objects and assigning it as an attribute to nodes in the graph, the dimension of the target graph is expanded from a single "physical connection" to "psychological state." This graph with emotional attributes not only records who is looking at whom in the meeting, but also records the current emotional tone of each participant (such as who is happy and who is angry). Therefore, it can perceive more refined information about the meeting and provide semantically rich underlying data support for subsequent understanding of the meeting situation.

[0077] Figure 5 Another flowchart illustrating a processing method according to an embodiment of the present disclosure is shown schematically.

[0078] like Figure 5 As shown, based on the foregoing embodiments, operation S130 may include at least one of operations S510 to S530.

[0079] When operating S510, the atmosphere information of a meeting event is determined based on relational information. The atmosphere information of a meeting event refers to a macroscopic description of the overall atmosphere or local interactive nature of the meeting. Unlike the emotions of a single object (such as someone being happy), the emotional information of a meeting event is a collective psychological state presented by the set of relational information between objects (i.e., the edges and their weights in the graph).

[0080] In operating the S520, attention center information for multiple objects is determined based on relational information. Attention center information indicates who is the visual focus or the core of attention convergence at the current moment of the meeting. In graph topology, this typically corresponds to nodes with high in-degree or strong centrality. By analyzing the directional nature of edges in the graph, the target node that is commonly pointed to by other nodes is identified, thereby determining the current speaker, the person being questioned, or a sudden source of distraction.

[0081] When operating the S530, potential emotional information of multiple objects is predicted based on relationship information. Potential emotional information is a prediction or trend analysis of the future direction of the meeting. The emotional state that may occur in the next moment can be inferred by analyzing the rate of change of interaction weights (such as whether the negative weight is increasing rapidly) or the evolution pattern of the graph structure (such as whether two opposing groups are forming).

[0082] For example, if the current situation is generally calm, but negative relational information about object A is gradually increasing, the predicted potential emotional information is "the risk of an impending conflict with object A"; or, if no new relational information appears for a long time or even if the relational information gradually decreases, the predicted potential emotional information is "the meeting is about to enter a period of fatigue or inattention".

[0083] According to embodiments of this disclosure, through in-depth mining of relationship information in the graph, multi-dimensional meeting insights can be output. Identification of the "center of attention" can assist in automatic directing or highlighting key points in meeting minutes; judgment of the "overall mood" can help the host control the pace of the meeting in real time; and prediction of "potential mood" can provide early warning of abnormal interactions.

[0084] Based on the foregoing embodiments, operation S510 may include: determining the weight corresponding to the relationship information based on the attributes of the nodes and the relationship information of the objects; if the attributes of the first node represent the positive emotions of the first object, the weight corresponding to the relationship information of the first object is positive; if the attributes of the first node represent the negative emotions of the first object, the weight corresponding to the relationship information of the first object is negative; and determining the sum of the weights corresponding to the relationship information of multiple objects as the atmosphere information of the meeting event.

[0085] A node's attributes are not merely simple labels, but a composite data structure containing multi-dimensional emotional features to accurately describe the psychological state of the object represented by the node. These attributes may include, but are not limited to: emotional tendency (e.g., positive, negative, neutral), emotional type (e.g., happy, angry, surprised, confused), and emotional intensity (e.g., intensity value or level).

[0086] For example, if the first object exhibits a laughing behavior, the attributes of the first node are set to a composite structure: {Emotional Tendency: Positive, Emotional Type: Happy, Emotional Level: 0.9 (High)}; if the second object exhibits a furrowed brow and pursed lips behavior, the attributes of the second node are set to: {Emotional Tendency: Negative, Emotional Type: Questioning / Angry, Emotional Level: 0.6 (Medium)}; if the third object exhibits calm listening, the attributes are set to: {Emotional Tendency: Neutral, Emotional Type: Calm, Emotional Level: 0.0}.

[0087] Weights are numerical indicators assigned to "edges" in a graph, used to quantify the nature and strength of the influence of relational information. The positive or negative polarity of a weight is determined by the emotional tendency of the object (source node) initiating the interaction, while the absolute value of the weight is determined by the degree (intensity) of that emotion. This means that the same act of "gazing" can have drastically different social meanings depending on the emotion of the "gazer."

[0088] Specifically, positive weights represent a "constructive" or "affinity" connection (such as agreement or appreciation); negative weights represent a "confrontational" or "repulsive" connection (such as opposition or questioning); and zero weights represent a "neutral" connection (such as simple information acquisition or observation). Furthermore, the more intense the emotional expression, the larger the absolute value of the corresponding weight. Additionally, when relational information points to itself or exhibits a same-sex relationship, its weight is reset to zero to avoid logical loops or invalid calculations.

[0089] For example, if object A is looking at object B, and the attribute of the node corresponding to object A is {emotional tendency: positive, emotional intensity: 0.3 (low)} (such as smiling), then the weight of the edge from the node corresponding to object A to the node corresponding to object B is set to a small positive value (such as +0.3); if the attribute is {emotional tendency: positive, emotional intensity: 0.9 (high)} (such as laughing and nodding), then the weight is set to a larger positive value (such as +0.9), representing a "strong positive association".

[0090] For example, if object C is looking at object B, and the attribute of the node corresponding to object C is {emotional tendency: negative, emotional intensity: 0.3 (low)} (e.g., slightly frowning), then the weight of the edge pointing from the node corresponding to object C to the node corresponding to object B is set to a small negative value (e.g., -0.3); if the attribute is {emotional tendency: negative, emotional intensity: 0.9 (high)} (e.g., slamming the table in anger), then the weight is set to a larger negative value (e.g., -0.9), representing a "strong negative association".

[0091] For example, if object D is gazing at object B, and the attribute of the node corresponding to object D is {emotional tendency: neutral} (such as expressionless), then the weight is set to 0, representing "objective attention".

[0092] The weights of all valid edges in the graph are summed to obtain a total score. This total score reflects the "emotional net value" within the meeting room. The accumulation of positive weights represents the superposition of positive interactions, while the accumulation of negative weights represents the superposition of negative interactions. The sign of the sum represents the overall nature, and the absolute value represents the intensity. For example, suppose there are three edges in the graph: edge 1 has a weight of +0.8 (A agrees with B), edge 2 has a weight of +0.5 (C pays attention to B), and edge 3 has a weight of -0.3 (D questions A). Then the emotional information of the meeting event is calculated as: 0.8 + 0.5 + (-0.3) = +1.0. This positive result indicates that the overall atmosphere of the meeting is positive.

[0093] In another embodiment, the emotional information of the meeting event can also be obtained based on the statistical values ​​of node attributes. The emotional attribute values ​​of all nodes in the graph can be extracted, and the average of these values ​​can be calculated. This average is used as the emotional information reflecting the overall tone of the meeting. Furthermore, due to the differences in the importance of participants' identities, a weighted average method can be used, pre-assigning an identity weight to each object (e.g., the meeting host or decision-maker has a higher identity weight, while ordinary participants have a lower identity weight). When calculating the overall emotional state, the emotional value of each object is multiplied by its corresponding identity weight, summed, and then divided by the total. For example, assuming there is object A (manager, identity weight = 2.0, emotional value = +0.8) and object B (employee, identity weight = 1.0, emotional value = -0.2) in the meeting, the overall emotional information of the meeting is calculated as: (0.8 × 2.0 + (-0.2) × 1.0) / (2.0 + 1.0) ≈ +0.7. This calculation method ensures that the emotions of key figures have a greater impact on the assessment of the overall meeting atmosphere.

[0094] Based on the foregoing embodiments, operation S520 may include: in response to the number of edges pointing to the third node being greater than or equal to a first threshold, determining the third object corresponding to the third node as the first focus. The number of edges pointing to the third node (i.e., the in-degree of the node) represents how many other objects are following the third object. When the number of followers reaches a certain scale (greater than or equal to the first threshold), the object is considered the current focus. Multiple first focus centers can exist; that is, any object with more followers than the first threshold can be considered a focus center. Alternatively, there can be only one first focus center; when multiple nodes have an in-degree greater than or equal to the first threshold, the node with the highest in-degree is designated as the first focus center.

[0095] Similarly, isolated centers in a meeting event can be identified using a method similar to that used for identifying centers of interest. The difference is that nodes with an in-degree less than a certain threshold, or nodes with the smallest in-degree, are considered isolated centers. This will not be elaborated further here.

[0096] Based on the foregoing embodiments, operation S530 may include at least one of the following: a first potential emotion prediction operation, a second potential emotion prediction operation, and a third potential emotion prediction operation.

[0097] The first potential sentiment prediction operation, in response to the number of edges pointing to the fourth node being greater than or equal to the third threshold, and the number of fifth nodes being greater than or equal to the fourth threshold, predicts the potential sentiment information of the fifth node as consensus sentiment. Here, the fifth node is a node connected to the fourth node by edges, and its attribute is characterized as positive or neutral sentiment. The fourth node typically represents the current focus figure. The fifth node represents the audience who are paying attention to the focus figure with a positive or objective attitude. When the focus figure receives enough attention (satisfying the third threshold), and the vast majority of these attendees exhibit a positive or neutral attitude (satisfying the fourth threshold), the system determines that the current meeting has reached "consensus."

[0098] For example, subject A is speaking (node ​​4). There are 10 people in the meeting, 8 of whom are looking at A (satisfying the third threshold). Of these 8 people, 7 are smiling or listening calmly (attribute is positive or neutral), which meets the fourth threshold. Based on this, it is predicted that the current sentiment is "consensus," meaning that subject A's proposal is likely to pass.

[0099] The second latent emotion prediction operation, in response to the presence of at least one target edge in the target graph, predicts the latent emotional information of multiple objects as "factional conflict emotion," where the attribute of at least one node connected to the target edge is characterized as negative emotion. This operation is used to capture explicit, point-to-point direct conflict. A conflict warning is triggered when interactions involving nodes with negative emotional attributes appear in the graph (e.g., angry stares, disgusted eye contact). For example, if an edge (target edge) from node B to node A is detected in the graph, and node B's attribute is displayed as "anger" (negative emotion), based on this hostile edge, the latent emotional information is predicted as "factional conflict emotion" or "localized confrontation."

[0100] The third latent sentiment prediction operation, in response to the number of edges pointing to the sixth node being greater than or equal to the fifth threshold, and the number of seventh nodes being greater than or equal to the sixth threshold, predicts the latent sentiment information of the seventh node as camp conflict sentiment. The seventh node is a node connected to the sixth node by an edge and whose attribute is characterized as negative sentiment, and / or the seventh node is a node not connected to the sixth node. This third latent sentiment prediction operation is used to identify complex "group splits," such as one group supporting / paying attention while another group opposes / ignores it (the seventh node). When the number of followers reaches a certain scale (satisfying the fifth threshold), and simultaneously the number of opponents or those who ignore the situation also reaches a certain scale (satisfying the sixth threshold), it indicates that the meeting room has split into two camps, or that the focal point has been resisted by some.

[0101] For example, subject A is speaking (node ​​six). Five people in the crowd are paying positive attention to A; however, at the same time, four other people in the crowd (node ​​seven) are either staring angrily at A (negative attribute) or completely ignoring A and playing on their phones (not connected to node six). Since the number of people on both sides has reached the threshold, it is predicted that there is currently "campal conflict" or "disagreement".

[0102] According to embodiments of this disclosure, through in-depth statistical analysis and logical judgment of graph data, it is possible to distinguish between "heated discussion" (consensus sentiment) and "intense arguments" (factional conflict). In particular, by introducing classification and counting of "supporters" (sixth node) and "opponents / ignorers" (seventh node), and combining multi-dimensional sentiment calculation with identity weights, it is possible to keenly capture implicit factional divisions and social isolation in meetings, thereby providing meeting managers with a more accurate situation assessment than intuition.

[0103] Figure 6 Another flowchart illustrating a processing method according to an embodiment of the present disclosure is shown schematically.

[0104] like Figure 6 As shown, based on the foregoing embodiments, the processing method may further include operation S610.

[0105] When operating the S610, target images corresponding to target information are presented in the environment of the meeting event or in the interactive window based on augmented reality technology. The target images are used to indicate at least one of the following: atmosphere information of the meeting event, focus information of multiple objects, and potential emotional information of multiple objects. The target image information is the transformation of abstract meeting analysis results (indication information) into visual graphic rendering data, defining the shape, color, position, transparency, and dynamic effects of the virtual elements that need to be visually presented.

[0106] In the context of a meeting or event, or within an interactive window, target images corresponding to target information are presented using augmented reality (AR) technology. Augmented reality is a display technology that overlays virtual information onto a real physical or virtual digital environment. It is used to spatially register and fuse generated virtual visual elements with the real-world scene or video footage seen by the user in real time. This can be achieved through wearable display devices (such as AR glasses and VR headsets) in a holographic overlay within the user's direct field of vision; or through flat-panel display devices (such as mobile phone screens and computer monitors) in a digital overlay on video streams captured by cameras.

[0107] According to embodiments of this disclosure, by utilizing augmented reality or virtual reality technology, the present disclosure transforms abstract meeting analysis results (such as emotional atmosphere and social structure) into an intuitive and immersive visual experience, enabling users to directly "see" the interpersonal network and emotional flow in the meeting without having to painstakingly interpret tedious data reports, thereby improving the efficiency of information perception and the intuitiveness of interaction.

[0108] Figure 7 The illustration schematically shows the presentation of target image information in an interactive window according to an embodiment of the present disclosure.

[0109] like Figure 7 As shown, visual elements generated from indicator information can be overlaid on or displayed in the sidebar of an interactive window that displays meeting events (such as a video conferencing interface). It should be noted that... Figure 7 The system displays multiple display methods, but in practical applications, users can choose one method to present the content individually or combine multiple methods according to their settings or system configuration.

[0110] Specifically, Figure 7 The enlarged view in the upper right corner shows a radar chart presentation. In this view, the current speaker (e.g., object A) is placed at the center of the visualization. Other objects (e.g., objects B, C, D, E, F, G) are arranged radially around the central object A. Relationships between objects are represented by connecting lines (e.g., arrows). To visually represent the nature of these relationships (i.e., the weights and sizes of the edges), the connecting lines are configured with different visual attributes (e.g., color, line style, thickness, or fill pattern). For example, connecting lines from objects C, E, and F to object A are green, representing positive attention (e.g., support, consensus); connecting lines from object B to object A are red, representing negative attention (e.g., opposition, conflict); connecting lines from objects D and G to object A are blue, representing neutral attention. Furthermore, in some embodiments, if an object (e.g., object G) does not have a connection to object A, the connecting line may not be rendered to reduce visual clutter.

[0111] Figure 7The enlarged view in the lower left corner shows a presentation based on a status display block (or dashboard) to represent the overall mood of a meeting event. As an example, the block can display different colors to represent different moods. For instance, green represents an overall "consensus" mood, blue represents a "neutral" mood, and red represents a "conflict" mood. The shade (saturation or transparency) of the color area can be further used to represent the intensity of the mood (e.g., darker colors represent more intense conflict). As another example, the block can be configured as a linear scale or a slider control. The two ends of the scale represent opposing mood polarities (e.g., left for "conflict," right for "consensus," and middle for "neutrality"), and a dynamic position cursor (such as a triangular arrow) indicates the current position of the overall meeting mood on the scale, providing real-time feedback on the meeting's atmosphere.

[0112] Figure 7 The magnified view in the lower right corner shows a text-label-based presentation method. Instructions can be directly converted into natural language text for display. For example, displaying "Central Person: Object A" in a specific area of ​​the screen (such as the bottom banner) clearly indicates the current focus; or displaying status descriptions such as "Consensus is being formed" or "Conflict detected" to help participants quickly understand the current meeting situation.

[0113] Figure 8 The diagram illustrates an interface that presents the emotional information of a single object in an interactive window according to an embodiment of the present disclosure.

[0114] like Figure 8 As shown, an interactive window (such as a video conferencing interface) contains multiple sub-windows for displaying the video feeds of participants. To visually indicate the current psychological state of each participant, an emotional state indicator (or display block) is displayed in the area associated with each participant's video window. Figure 8 In the example shown, the emotion status indicator is presented as a circular icon located in a corner of the video window (e.g., the upper right corner). After determining the emotion information of the object, the visual attributes of the indicator (e.g., color, brightness, or flashing frequency) are dynamically adjusted. For example, using color as a distinguishing dimension, when an object's node attribute is determined to represent a negative emotion (e.g., anger, aversion), the corresponding display block is presented in the first color (e.g., red) to serve as a warning; when an object's node attribute is determined to represent a neutral emotion (e.g., calm, thinking), the corresponding display block is presented in the second color (e.g., blue) to represent a normal state; and when an object's node attribute is determined to represent a positive emotion (e.g., happiness, agreement), the corresponding display block is presented in the third color (e.g., green) to represent a positive state.

[0115] It should be noted that, although Figure 8 The example shown uses a circular icon to indicate emotional state, but this is merely illustrative. In practical applications, the presentation of target image information can take many forms. For example, the emotional state indicator can be rendered as the border of a video window, with the emotion indicated by changing the border color (e.g., red or green); or, the emotional state indicator can be rendered as the background color or a semi-transparent overlay of the video screen, creating a corresponding emotional atmosphere by changing the background hue; or, text labels or emoticons can be displayed directly next to the subject's avatar. According to embodiments of this disclosure, participants can readily perceive the emotional state of each member while browsing the video screen, thereby better adjusting their communication strategies.

[0116] Figure 9 Another flowchart illustrating a processing method according to an embodiment of the present disclosure is shown schematically.

[0117] like Figure 9 As shown, based on the aforementioned embodiments, the output information is the voice output information or text output information of the target object, and the processing method may further include operations S910 to S920.

[0118] In operation S910, in response to the presence of first output information in the output information, the number of objects satisfying the target condition is determined based on the relational information. The target condition represents a negative emotion indicated by the relational information. The first output information refers to the voice or text content output by an object (such as the current speaker) that is identified as potentially risky or non-compliant. The system is pre-loaded with a knowledge base containing prohibition rules and compliance rules. When the speech content monitored in real time matches a prohibition rule, the presence of first output information is determined. At this time, based on the relational information in the graph, it is determined whether other objects have shown negative emotional feedback (such as questioning or aversion) to the speaker, thereby determining the number of people with negative emotions.

[0119] For example, suppose the speaker says, "I guarantee that our revenue will grow by 50% next quarter!" (first output information). This content hits the rule of "prohibiting absolute performance commitments." At this point, querying the target graph reveals that there are 3 other objects (such as investors) watching the speaker, and their node attributes are all "doubt" or "frowning" (negative emotions). The total number of objects that meet the target condition is 3.

[0120] In operation S920, in response to the number of objects being greater than or equal to a threshold, a first alert instruction is generated and sent. This first alert instruction is used to alert the target objects, and the first output information has target semantics. The first alert instruction is a real-time warning signal designed to interrupt or remind the target objects of the current speaking risk. The target semantics refer to the specific risk semantic content violated by the speaking content. When negative feedback from listeners reaches a certain scale (the threshold), it is confirmed that the risky speech has caused adverse consequences, thus triggering an alarm. The alert method can be multi-channel, such as visual highlighting, tactile vibration, or audio whispering.

[0121] For example, a threshold of 2 is set. Since the number of negative feedback responses in the previous example is 3 (greater than 2), a first prompt instruction is immediately generated. This instruction controls the speaker's wristband to vibrate at a specific frequency, or flashes red text on the speaker's AR glasses / teleprompter screen: "Warning: Absolute commitment risk detected (target semantics), please correct immediately!"

[0122] In some embodiments, the processing method may further include: in response to the presence of second output information in the output information, generating and sending a second instruction, wherein the second instruction is used to stop the prompting action corresponding to the first instruction, and the second output information has a negative semantic meaning regarding the target semantics. The second output information refers to supplementary explanations, corrective statements, or disclaimers made by the object after receiving the warning. The negative semantic meaning regarding the target semantics means that the supplementary statement eliminates the previous risk, bringing the overall expression back within the compliant range. When compliant remedial measures are detected, a second instruction is sent to deactivate the alarm, forming a closed-loop risk control process.

[0123] For example, after feeling the bracelet vibrate, the spokesperson immediately added, "...Of course, this is a forward-looking forecast based on the current market, and the actual results may differ due to factors such as supply chain risks." (Second output information). Recognizing that this content negates the aforementioned absolute commitment, eliminating the risk of the previous absolute commitment, a second instruction is sent to stop the bracelet's vibration and the red flashing of the screen, and may display a green "compliance" icon.

[0124] According to embodiments of this disclosure, by combining content detection from a knowledge base containing prohibitive and compliance rules with emotional feedback from "graph interaction," the audience's real-time reactions are introduced as a weight for risk confirmation, avoiding false alarms for jokes or rhetorical devices. Simultaneously, through a closed-loop feedback system of "alert-correction-cancellation," the speaker can be assisted in correcting slips of the tongue or imprecise expressions in real time when a risk occurs (rather than after-the-fact accountability), thereby effectively reducing legal risks or communication incidents during meetings.

[0125] Figure 10A block diagram of a processing apparatus according to an embodiment of the present disclosure is shown schematically.

[0126] like Figure 10 As shown, the processing device 1000 may include a first acquisition module 1010, a first determination module 1020, and a first generation module 1030.

[0127] The first acquisition module 1010 is used to acquire target data of the meeting event, which indicates multiple objects of the meeting event. In some embodiments, the first acquisition module 1010 may be used to perform operation S110 in the above processing method, which will not be described in detail here.

[0128] The first determining module 1020 is used to determine the relationship information of the object based on the target data. The relationship information represents the emotional or actional relationship of the object with respect to at least one of multiple objects. In some embodiments, the first determining module 1020 can be used to perform operation S120 in the above processing method, which will not be described in detail here.

[0129] The first generation module 1030 is used to generate target information for a meeting event based on objects and their relationship information. The target information indicates the interaction state in the meeting event. In some embodiments, the first generation module 1030 can be used to perform operation S130 in the above processing method, which will not be described in detail here.

[0130] Any one or more of the modules, submodules, units, and subunits according to embodiments of the present disclosure, or at least part of the functions of any one or more of them, can be implemented in one module. Any one or more of the modules, submodules, units, and subunits according to embodiments of the present disclosure can be implemented by dividing them into multiple modules. Any one or more of the modules, submodules, units, and subunits according to embodiments of the present disclosure can be at least partially implemented as hardware circuitry, such as a Field-Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a System-on-Chip, a System-on-a-Substrate, a System-on-Package, an Application-Specific Integrated Circuit (ASIC), or implemented in hardware or firmware by any other reasonable means of integrating or packaging circuitry, or implemented in software, hardware, or firmware, or in any suitable combination of any of these three implementation methods. Alternatively, one or more of the modules, submodules, units, and subunits according to embodiments of the present disclosure can be at least partially implemented as computer program modules, which, when run, can perform corresponding functions.

[0131] For example, any plurality of the first acquisition module 1010, the first determination module 1020, and the first generation module 1030 can be combined into one module / unit / subunit, or any one of these modules / units / subunits can be split into multiple modules / units / subunits. Alternatively, at least part of the functionality of one or more of these modules / units / subunits can be combined with at least part of the functionality of other modules / units / subunits and implemented in one module / unit / subunit. According to embodiments of this disclosure, at least one of the first acquisition module 1010, the first determination module 1020, and the first generation module 1030 can be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or implemented in hardware or firmware by any other reasonable means of integrating or packaging the circuitry, or implemented in any one of the three implementation methods of software, hardware, and firmware, or in a suitable combination of any of these. Alternatively, at least one of the first acquisition module 1010, the first determination module 1020, and the first generation module 1030 may be implemented at least partially as a computer program module, which can perform corresponding functions when the computer program module is run.

[0132] It should be noted that the data processing system part in the embodiments of this disclosure corresponds to the data processing method part in the embodiments of this disclosure. The specific description of the data processing system part is referred to in the data processing method part, and will not be repeated here.

[0133] Figure 11 A block diagram of an electronic device suitable for implementing the methods described above, according to embodiments of the present disclosure, is illustrated schematically. Figure 11 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0134] like Figure 11 As shown, an electronic device 1100 according to an embodiment of the present disclosure includes a processor 1101, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1102 or a program loaded from a storage portion 1108 into a random access memory (RAM) 1103. The processor 1101 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 1101 may also include onboard memory for caching purposes. The processor 1101 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0135] RAM 1103 stores various programs and data required for the operation of electronic device 1100. Processor 1101, ROM 1102, and RAM 1103 are interconnected via bus 1104. Processor 1101 performs various operations of the method flow according to embodiments of the present disclosure by executing programs in ROM 1102 and / or RAM 1103. It should be noted that the programs may also be stored in one or more memories other than ROM 1102 and RAM 1103. Processor 1101 may also perform various operations of the method flow according to embodiments of the present disclosure by executing programs stored in said one or more memories.

[0136] According to embodiments of this disclosure, the electronic device 1100 may further include an input / output (I / O) interface 1105, which is also connected to a bus 1104. The electronic device 1100 may also include one or more of the following components connected to the input / output (I / O) interface 1105: an input section 1106 including a keyboard, mouse, etc.; an output section 1107 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1108 including a hard disk, etc.; and a communication section 1109 including a network interface card such as a LAN card, modem, etc. The communication section 1109 performs communication processing via a network such as the Internet. A drive 1110 is also connected to the input / output (I / O) interface 1105 as needed. A removable medium 1111, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 1110 as needed so that computer programs read from it can be installed into the storage section 1108 as needed.

[0137] According to embodiments of this disclosure, the method flow according to embodiments of this disclosure can be implemented as a computer software program. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable storage medium, the computer program containing program code for performing the methods shown in the flowchart. In such embodiments, the computer program can be downloaded and installed from a network via communication section 1109, and / or installed from removable medium 1111. When the computer program is executed by processor 1101, it performs the functions defined in the system of embodiments of this disclosure. According to embodiments of this disclosure, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0138] This disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs that, when executed, implement the method according to the embodiments of this disclosure.

[0139] According to embodiments of this disclosure, the computer-readable storage medium can be a non-volatile computer-readable storage medium. Examples include, but are not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this disclosure, the computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0140] For example, according to embodiments of this disclosure, a computer-readable storage medium may include one or more memories other than the ROM 1102 and / or RAM 1103 described above and / or ROM 1102 and RAM 1103.

[0141] Embodiments of this disclosure also include a computer program product comprising a computer program containing program code for performing the methods provided in the embodiments of this disclosure. When the computer program product is run on an electronic device, the program code is used to enable the electronic device to implement the processing methods provided in the embodiments of this disclosure.

[0142] When the computer program is executed by the processor 1101, it performs the functions defined in the system / apparatus of this disclosure embodiments. According to embodiments of this disclosure, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0143] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices or magnetic storage devices. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and downloaded and installed via communication section 1109, and / or installed from removable medium 1111. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof. According to embodiments of this disclosure, program code for executing the computer programs provided in embodiments of this disclosure can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code may be executed entirely on a user computing device, partially on a user device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing devices can be connected to user computing devices via any type of network, including local area networks (LANs) or wide area networks (WANs), or they can be connected to external computing devices (e.g., via the Internet using an Internet service provider).

[0144] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions. Those skilled in the art will understand that the features described in the various embodiments of the present disclosure can be combined and / or combined in various ways, even if such combinations are not explicitly described in the present disclosure. In particular, the features described in the various embodiments of this disclosure may be combined and / or combined in various ways without departing from the spirit and teachings of this disclosure. All such combinations and / or combinations fall within the scope of this disclosure.

[0145] The embodiments of this disclosure have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of this disclosure. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this disclosure, and all such substitutions and modifications should fall within the scope of this disclosure.

Claims

1. A processing method, comprising: Retrieve target data from multiple objects in a meeting event; Based on the target data, relationship information between the multiple objects is determined, wherein the relationship information represents the emotional or behavioral relationship between any one of the multiple objects and other objects; Based on the relationship information between the multiple objects, target information for the meeting event is generated, and the target information is used to display the interaction status in the meeting event.

2. The method according to claim 1, wherein generating target information for the meeting event based on the relationship information between the plurality of objects includes: Based on the relationship information between the multiple objects, a target map is generated; The target information of the conference event includes the target map; The target map includes: Nodes are used to represent the objects; Edges are used to represent the relationship information between the multiple objects.

3. The method according to claim 2, wherein the target data includes: At least one of the following regarding the meeting event: text data, audio data, image data, video data, and interactive data; Based on the relationship information between the multiple objects, a target map is generated, including: Multiple objects are identified as multiple nodes in the target graph; Based on the target data, determine the emotional information and / or action information of the multiple objects; Based on the emotional information and / or action information of the first object among the plurality of objects, a first edge of the first node corresponding to the first object is determined, and the edge is used to indicate the relationship information between the first object and other objects. The target graph is generated based on at least one first node and at least one first edge.

4. The method according to claim 3, wherein the action information includes: At least one of the object's gaze, head posture, body posture, and movement; Based on the emotion information and / or action information of a first object among the plurality of objects, determine the first edge of the first node corresponding to the first object, including: Based on the action information, determine the attention vector of the first object; In response to the attention vector of the first object pointing to the second object among the plurality of objects, it is determined that the first node corresponding to the first object points to the first edge of the second node corresponding to the second object; Based on the emotional information, determine the attributes of the first node corresponding to the first object; The target graph is generated based on at least one first node and at least one first edge, including: The target graph is generated based on the attributes of at least one first node, the second node, and at least one first edge and the first node.

5. The method according to claim 4, wherein the meeting event includes a first meeting scenario and a second meeting scenario, the second meeting scenario is presented in the first meeting scenario through an interactive window, and the target data further includes interaction information between the object and the interactive window. In response to the attention vector of the first object pointing to the second object among the plurality of objects, determining that the first edge of the first node corresponding to the first object points to the first edge of the second node corresponding to the second object, further comprising: In response to the attention vector of the first object pointing to the second object in the interaction window, based on the attention vector and / or the interaction information, it is determined that the first edge of the first node corresponding to the first object points to the second node corresponding to the second object.

6. The method according to claim 1, wherein generating the target information of the meeting event based on the relationship information between multiple objects includes at least one of the following: The atmosphere information of the meeting event is determined based on the relationship information; Based on the relationship information, determine the attention center information of the multiple objects; Predict the potential emotional information of the multiple objects based on the relationship information.

7. The method according to claim 6, wherein determining the atmosphere information of the meeting event based on the relationship information comprises: Based on the attributes of the node and the relationship information of the object, determine the weight corresponding to the relationship information; If the attributes of the first node represent the positive sentiment of the first object, the weight corresponding to the relation information of the first object is positive. If the attribute of the first node represents the negative emotion of the first object, the weight corresponding to the relation information of the first object is negative; The sum of the weights corresponding to the relationship information of multiple objects is determined as the atmosphere information of the meeting event; The step of determining the attention center information of the multiple objects based on multiple relational information includes: In response to the number of edges pointing to the third node being greater than or equal to a first threshold, the third object corresponding to the third node is determined as the first center of interest; The prediction of potential emotional information of the multiple objects based on multiple relational information includes at least one of the following: In response to the number of edges pointing to the fourth node being greater than or equal to the third threshold and the number of fifth nodes being greater than or equal to the fourth threshold, the potential emotional information of the fifth node is predicted to be consensus emotion, wherein the fifth node is a node connected by an edge pointing to the fourth node and whose attribute is characterized as positive emotion or neutral emotion. In response to the presence of at least one target edge in the target graph, the potential emotional information of the plurality of objects is predicted to be camp conflict emotion, and the attribute of at least one node connected to the target edge is characterized as negative emotion; In response to the number of edges pointing to the sixth node being greater than or equal to the fifth threshold, and the number of seventh nodes being greater than or equal to the sixth threshold, the potential emotional information of the seventh node is predicted to be a camp conflict emotion, wherein the seventh node is a node connected to the sixth node by an edge and whose attribute is characterized as negative emotion, and / or the seventh node is a node not connected to the sixth node.

8. The method according to claim 1 or 4, further comprising: In the environment of the meeting event, or in the interactive window, a target image corresponding to the target information is presented based on augmented reality technology. The target image is used to indicate at least one of the atmosphere information of the meeting event, the focus information of the multiple objects, and the potential emotional information of the multiple objects.

9. The method according to claim 1, wherein the target data further includes output information of the target object, the output information being voice output information or text output information of the target object, and the method further includes: In response to the presence of first output information in the output information, the number of objects that meet the target conditions is determined based on the relationship information; The target condition indicates that the emotion indicated by the relational information is a negative emotion; In response to the number of objects being greater than or equal to a quantity threshold, a first prompt instruction is generated and sent. The first prompt instruction is used to prompt the target object, and the first output information has target semantics.

10. An electronic device, comprising: At least one processor; and a memory connected to at least one processor; wherein the memory stores instructions executable by at least one of the processors, the instructions being executed by at least one of the processors to enable at least one of the processors to perform at least one of the following operations: acquiring target data of multiple objects in a conference event; and determining relationship information between the multiple objects based on the target data, the relationship information representing the emotional or actional relationship between any one of the multiple objects and other objects; Based on the relationship information between multiple objects, target information for the meeting event is generated. This target information is used to display the interactive state during the meeting event.

Images