CN116850603A

CN116850603A - Game item recommending method and device, storage medium and electronic equipment

Info

Publication number: CN116850603A
Application number: CN202310894114.4A
Authority: CN
Inventors: 汪棋欣
Original assignee: Guangzhou Boguan Information Technology Co Ltd
Current assignee: Guangzhou Boguan Information Technology Co Ltd
Priority date: 2023-07-19
Filing date: 2023-07-19
Publication date: 2023-10-10

Abstract

The disclosure provides a game item recommending method, a game item recommending device, a computer readable storage medium and electronic equipment, and relates to the technical field of games and live broadcasting. The game item recommending method applied to the game server side of the target game comprises the following steps: acquiring eye movement data of a player wearing a head-mounted device in the target game, and generating game item recommendation information for the player according to the eye movement data; and sending the game item recommendation information to a main broadcasting end associated with the target game in a live broadcasting platform, so that the main broadcasting end generates a live broadcasting video for recommending the game item according to the game item recommendation information, and the live broadcasting video is used for being displayed to the player. The method and the device are beneficial to players to purchase the needed game objects efficiently and accurately, and improve game experience.

Description

Game item recommending method and device, storage medium and electronic equipment

Technical Field

The present disclosure relates to the field of gaming and live broadcasting technologies, and in particular, to a game item recommendation method, a game item recommendation device, a computer-readable storage medium, and an electronic apparatus.

Background

Currently, in most games, the information available to a player is limited when the player wants to purchase a game item. May be embodied in: the variety of game items in a game store is large, and it is difficult for a player to conveniently find a suitable or interesting game item therein; the introduction information about the game item is small, and the player cannot sufficiently know the game item; etc. Thus, it is difficult for the player to purchase the desired game item efficiently and accurately, affecting the game experience.

Disclosure of Invention

The present disclosure provides a game item recommending method, a game item recommending apparatus, a computer-readable storage medium, and an electronic device to solve, at least to some extent, a problem that it is difficult for a player to purchase a desired game item efficiently and accurately in a game.

According to a first aspect of the present disclosure, there is provided a game item recommendation method applied to a game server of a target game; the method comprises the following steps: acquiring eye movement data of a player wearing a head-mounted device in the target game, and generating game item recommendation information for the player according to the eye movement data; and sending the game item recommendation information to a main broadcasting end associated with the target game in a live broadcasting platform, so that the main broadcasting end generates a live broadcasting video for recommending the game item according to the game item recommendation information, and the live broadcasting video is used for being displayed to the player.

According to a second aspect of the present disclosure, there is provided a game item recommendation method applied to a hosting end associated with a target game in a live platform; the method comprises the following steps: acquiring game item recommendation information for a player in the target game; the game item recommendation information is generated from eye movement data of the player wearing a head mounted device in the target game; and generating a live video for recommending the game item according to the game item recommendation information, wherein the live video is used for being displayed to the player.

According to a third aspect of the present disclosure, there is provided a game item recommendation apparatus applied to a game server of a target game; the device comprises: an eye movement data processing module configured to acquire eye movement data of a player wearing a head-mounted device in the target game, and generate game item recommendation information for the player according to the eye movement data; the game item recommendation information sending module is configured to send the game item recommendation information to a main broadcasting end associated with the target game in a live broadcasting platform, so that the main broadcasting end generates a live broadcasting video for recommending the game item according to the game item recommendation information, and the live broadcasting video is used for being displayed to the player.

According to a fourth aspect of the present disclosure, there is provided a game item recommendation device applied to a hosting end associated with a target game in a live platform; the device comprises: a game item recommendation information acquisition module configured to acquire game item recommendation information for a player in the target game; the game item recommendation information is generated from eye movement data of the player wearing a head mounted device in the target game; and the live video generation module is configured to generate live video for recommending the game item according to the game item recommendation information, wherein the live video is used for being shown to the player.

According to a fifth aspect of the present disclosure, there is provided a computer readable storage medium having stored thereon a computer program which when executed by a processor implements the game item recommendation method of the first or second aspect described above and possible implementations thereof.

According to a sixth aspect of the present disclosure, there is provided an electronic device comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform the game item recommendation method of the first or second aspect described above and possible implementations thereof via execution of the executable instructions.

The technical scheme of the present disclosure has the following beneficial effects:

according to the scheme, personalized and targeted game item recommendation information can be generated for the player, and game item recommendation is conducted for the player in a live broadcast mode, so that the player does not need to search for needed or interested items from a large number of game items, and a live broadcast video can provide detailed game item introduction information for the player, so that the player can purchase the needed game items efficiently and accurately, and game experience is improved. On the other hand, game item recommendation information is generated according to eye movement data of the head-mounted device worn by the player in the game, and the eye movement data can reflect the sight direction of the player, so that game items required or interested by the player can be accurately estimated, the generated game item recommendation information is matched with the player, and the efficiency and experience of purchasing the game items by the player can be improved.

Drawings

FIG. 1 illustrates a system architecture diagram of an operating environment of the present exemplary embodiment;

FIG. 2 is a flowchart of a game item recommendation method performed by a game server in the present exemplary embodiment;

FIG. 3 illustrates a flowchart of one method of generating game item recommendation information in the present exemplary embodiment;

FIG. 4 illustrates another flow chart for generating game item recommendation information in the present exemplary embodiment;

FIG. 5 illustrates a flowchart of a game item recommendation method performed by a host end in the present exemplary embodiment;

fig. 6 shows a flowchart of determining a virtual scene in the present exemplary embodiment;

fig. 7 is a schematic view showing a structure of a game item recommending apparatus in the present exemplary embodiment;

FIG. 8 is a schematic view showing the structure of another game item recommender in the present exemplary embodiment;

fig. 9 shows a schematic structural diagram of an electronic device in the present exemplary embodiment.

Detailed Description

Exemplary embodiments of the present disclosure will be described more fully hereinafter with reference to the accompanying drawings.

The drawings are schematic illustrations of the present disclosure and are not necessarily drawn to scale. Some of the block diagrams shown in the figures may be functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, or in hardware modules or integrated circuits, or in networks, processors or microcontrollers. Embodiments may be embodied in many forms and should not be construed as limited to the examples set forth herein. The described features, structures, or characteristics of the disclosure may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough description of embodiments of the present disclosure. However, it will be recognized by one skilled in the art that one or more of the specific details may be omitted, or other methods, components, devices, steps, etc. may be used instead of one or more of the specific details in implementing the aspects of the present disclosure.

In most games, when a player wants to purchase a game item, the player needs to browse through a game store (or other game transaction interface) and find the desired game item. However, game shops typically present all kinds of game items and have little introduction information about the game items, such as only a two-sentence text, so that players cannot fully understand each game item, and it is difficult to conveniently find a suitable or interesting game item. Thus, it is difficult for the player to purchase the desired game item efficiently and accurately, affecting the game experience.

In view of the above, exemplary embodiments of the present disclosure provide a game item recommending method for recommending game items in a target game to a player in the target game, capable of providing rich and targeted game item information so that the player purchases a desired game item efficiently and accurately.

Fig. 1 shows a system architecture diagram of an operating environment of the present exemplary embodiment. The system architecture 100 may include a game system 110 and a live system 120 for a target game. The target game is a game that supports a head mounted device, i.e., an operation by which a player can play the target game with the head mounted device, including, but not limited to, VR (Virtual Reality) glasses, AR (Augmented Reality) glasses, MR (Mixed Reality) head mounted devices, and the like. Gaming system 110 may include a game server 1101 and a game client 1102. Game client 1102 refers to a client program of a target game, and may also refer to a terminal device running the client program. The terminal device may be the above-mentioned head-mounted device, or a non-head-mounted terminal device such as a personal computer, a smart phone, a game machine, or the like. If the game client 1102 is running on a non-head-mounted terminal device, the head-mounted device may be used as a companion peripheral. The game server 1101 is a background system that provides a target game service, and may be a server, or a cluster formed by a plurality of servers. In one embodiment, game server 1101 may include a game processing server, a game item transaction server, etc., and different servers may perform different functions.

The live system 120 may include a live service side 1201, a main cast side 1202, and a viewer side 1203. Wherein the anchor 1202 is the terminal used by the anchor to turn on the live broadcast. In one embodiment, the presenter side 1202 may be configured with a camera for capturing a presenter's view, such as capturing a video stream having a presenter's face as the primary content for display within a live room. The audience terminal 1203 refers to a terminal used by an audience entering a living room and watching living broadcast, and can comprise a mobile phone, a personal computer, a tablet personal computer, intelligent wearable equipment, an intelligent television and the like. Live service 1201 refers to a background system that provides live services.

The anchor side 1202 may form a live video stream by capturing an anchor portrait in real time, capturing an anchor side 1202 screen in real time, and the like, and may also form a live audio stream from the equipment audio data of the anchor side 1202, the environment audio data collected in real time, and the like, and send the live video stream and the live audio stream to the live server side 1201. The live broadcast server 1201 sends or processes the live video stream and the live audio stream (e.g., adjusts parameters such as size and definition of the live video stream, and encodes and decodes the live video stream and the live audio stream) and then sends the live video stream and the live audio stream to the corresponding viewer 1203, so that the viewer 1203 can present live broadcast pictures and live broadcast sounds. In one embodiment, the anchor 1202 may adopt a virtual live broadcast manner, and the anchor 1202 or the live broadcast server 1201 may use an Engine such as a UE (universal Engine) to render a virtual scene, and fuse the virtual scene with the real-time shot or real-time intercepted picture to form a live broadcast picture. In one embodiment, spectator 1203 and game client 1102 may include the same terminal device, i.e., one terminal device running both the target game and watching the live.

In one embodiment, the system architecture 100 may further include a headset server 130 for providing background services to the headset. For example, a player may choose to turn on a target game on a headset, the headset sending a game on instruction and player information to the headset server 130, the headset server 130 interacting with the game server 1101 of the target game according to the game on instruction, such that the game server 1101 can provide target game services to the headset (or game client 1102).

The connection between the game server 1101 and the game client 1102, the connection between the game server 1101 and the live broadcast server 1201, the connection between the live broadcast server 1201 and the anchor 1202, and the connection between the head-mounted device server 130 and the game server 1101 can be through a wired or wireless communication link for data transmission.

In one embodiment, gaming system 110 may be a cloud gaming system. Cloud gaming refers to a game style based on cloud computing. In the cloud game operation mode, the game program operation main body and the game picture presentation main body are separated, the storage and operation of the control and interaction method in the game are completed on a cloud game server (such as the game server 1101), and the function of a cloud game client (such as the game client 1102) is used for receiving and sending data and presenting the game picture. For example, the cloud game client may be a display device with a data transmission function near the user side, such as a mobile terminal, a television, a computer, a palm computer, etc.; and the cloud game server for information processing is a cloud game server. When playing a game, a player operates the cloud game client to send an operation instruction to the cloud game server, the cloud game server runs the game according to the operation instruction, codes and compresses data such as game pictures and the like, returns the data to the cloud game client through a network, and finally decodes the data through the cloud game client and outputs the game pictures.

The game item recommendation method in the present exemplary embodiment will be described below in terms of the game server 1101 and the anchor 1202, respectively.

FIG. 2 illustrates an exemplary flow of a game item recommendation method performed by the game server 1101, which may include the following steps S210 and S220:

step S210, acquiring eye movement data of the wearing head-mounted device of the player in the target game, and generating game item recommendation information for the player according to the eye movement data;

in step S220, the game item recommendation information is sent to the anchor 1202 associated with the target game in the live platform, so that the anchor 1202 generates a live video for recommending the game item according to the game item recommendation information, and the live video is used for showing to the player.

Based on the method shown in fig. 2, on one hand, a scheme for recommending game items for a player is provided, personalized and targeted game item recommending information can be generated for the player, and game item recommending is performed for the player in a live broadcast mode, so that the player does not need to search for needed or interested items from a large number of game items, and a live broadcast video can provide detailed game item introducing information for the player, thereby being beneficial to the player to purchase the needed game items efficiently and accurately and improving game experience. On the other hand, game item recommendation information is generated according to eye movement data of the head-mounted device worn by the player in the game, and the eye movement data can reflect the sight direction of the player, so that game items required or interested by the player can be accurately estimated, the generated game item recommendation information is matched with the player, and the efficiency and experience of purchasing the game items by the player can be improved.

Each step in fig. 2 is described in detail below.

Referring to fig. 2, in step S210, eye movement data of a player wearing a head-mounted device in a target game is acquired, and game item recommendation information for the player is generated from the eye movement data.

The player may be any player in the target game, or may be a specific player. For example, a function of recommending game items live may be set in the target game, and the player may be a player who has turned on the function. When a player wears the head-mounted device, an eye movement instrument (or a camera, an infrared sensor and the like) built in the head-mounted device can monitor changes of eyes of the player, so that eyeball (or pupil) position data can be obtained, and the eyeball position data can be analyzed to obtain line-of-sight direction data and the like. The eye movement data may include eyeball position data, gaze direction data, and the like. In one embodiment, the eye movement data may include three-dimensional coordinates of the eye position at different times, which may form an ordered set of three-dimensional coordinate sequences.

In this exemplary embodiment, the game server 1101 may obtain eye movement data of a player during a specific game course of a target game, which may reflect interactions between the player and game items. For example, eye movement data of a player in a game play may be acquired, or eye movement data of a player while viewing a game item-related interface may be acquired. The game server 1101 may obtain eye movement data from the game client 1102 or the headset server 130, which is not limited by this disclosure.

The eye movement data can characterize the line of sight of the player, whereby the player's interest or need for a particular game item can be determined, thereby generating game item recommendation information for the player.

In one embodiment, referring to fig. 3, the generating game item recommendation information for a player according to eye movement data may include the following steps S310 and S320:

step S310, determining the fixation of the player to the game picture in the game process according to the eye movement data.

The gazing game screen refers to a partial game screen that a player gazes on in the entire game screen that the player can see. According to the eye movement data, the fixation position of the player in the game picture can be determined, and then the game picture of the fixation position is extracted, so that the fixation game picture is obtained.

In one embodiment, the determining, according to the eye movement data, the gaze of the player in the game process may include the following steps:

determining a gaze point of the player during one or more unit times in the game based on the eye movement data;

and aggregating the fixation points in each unit time, determining the fixation area in each unit time according to the aggregation result, and intercepting the fixation game picture from the game picture by utilizing the fixation area.

The gaze point may be a two-dimensional point obtained by projecting three-dimensional coordinates of the eye position in the eye movement data onto a plane on which the game screen is located, and has two-dimensional coordinates on the plane on which the game screen is located. The gaze point may retain time information in the eye movement data, i.e. each gaze point may have a corresponding acquisition time. The unit time is the granularity of time for processing the eye movement data and can be determined empirically or according to specific needs, for example, can be 5 seconds. For example, the time period for collecting the player's eye movement data may be divided by one unit time every 5 seconds, and the gaze point may be corresponding to the corresponding unit time.

The gaze points per unit time are then aggregated. During play, the player's gaze direction does not change constantly, but typically remains in one orientation for a period of time, then moves to another orientation for a further period of time, and the gaze point may be more focused while the gaze remains. Therefore, the method can aggregate the gaze points which are close to each other in time and space, so that the number of the gaze points is reduced, and the processing efficiency is improved.

In one embodiment, the aggregating the gaze points in each unit time may include the following steps:

And aggregating a plurality of fixation points with time intervals smaller than a first preset time length and inter-coordinate distances smaller than the first preset distance in each unit time into one fixation point.

The first preset time length is a criterion for judging whether the plurality of gaze points are close in time, the first preset distance is a criterion for judging whether the plurality of gaze points are close in space, and both the first preset time length and the second preset time length can be determined according to experience or specific requirements. And (3) using T1 to represent a first preset time, A1 to represent a first preset distance, and clustering the fixation points in each unit time to obtain one or more fixation point classes, wherein the time interval between any two fixation points in each fixation point class is smaller than T1, and the distance between coordinates is smaller than A1. The gaze points in each gaze point class are then aggregated into one gaze point, e.g. the cluster center of each gaze point class may be used as the aggregated gaze point of the class.

Based on the result of the aggregation, a smaller number of gaze points per unit time are obtained, from which gaze areas are formed. For example, a bounding box (the bounding box may be rectangular or other polygonal shape) surrounding the gaze point may be used as the gaze area, or the gaze area may be obtained by enlarging the bounding box according to a certain ratio (for example, enlarging the side length of the bounding box to 1.2 times, etc.). And intercepting a local picture where the gazing area is located from the game picture to obtain the gazing game picture.

In one embodiment, after determining the gaze point of the player for one or more units of time during the game based on the eye movement data, outliers may be removed from the gaze point. For example, if there are two gaze points p that are adjacent (adjacent means chronologically adjacent) _i And p _i+1 The distance between the coordinates is larger than the second preset distance, and the two gaze points are a plurality of gaze points (p can be used _i Or p _i+1 N gaze points being the center) is less than a third predetermined distance, then it can be considered that at gaze point p _i And p _i+1 Where eye jump occurs, i.e. the player does not actually look at the gaze point p _i Or p _i+1 And the two ends where eye jump occurs are true gaze points. Thereby the point of gaze p _i And p _i+1 And eliminating the abnormal points. For another example, if there is a point of regard p _i Which is associated with two adjacent gaze points p _i-1 、p _i+1 The inter-coordinate distances of the two points are all larger than the fourth preset distance, and the point of gaze p is indicated _i As outliers, they can be culled as outliers. The values of n and the like of the second, third and fourth preset distances can be determined according to experience or specific requirements.

By eliminating abnormal points from the gaze point, the quality of the gaze point can be ensured, the accuracy of subsequent processing is improved, and the processing efficiency is improved.

Step S320, image recognition is carried out on the gazing game picture, and game item recommendation information for the player is generated according to the image recognition result.

The image recognition of the gazing game screen may be performed for the main purpose of recognizing the game object in the image, but the present disclosure is not limited to this, and may be performed for recognizing other props, skills, and the like related to the game object in the image. Image recognition may be achieved by pre-trained machine learning models (e.g., object detection networks), and the like. For example, a large number of sample images including or not including a specific object (such as a game item, prop, skill, etc. to be identified) may be extracted in advance from a game screen, and a machine learning model for image identification may be trained using the sample images and the tag data thereof, according to whether or not they include the specific object, the category of the specific object included, etc. to generate the tag data of the sample images. After training is completed, the gaze game screen may be image-identified using the trained model in step S320.

The image recognition result may include: whether a specific target is contained in the gazing game picture or not, and the category of the contained specific target. Thereby determining which game items, props or skills, etc. the player looks at more during the game, and the game items related to the game items can be used as potential game items which the player may need or be interested in, so as to generate game item recommendation information.

In one embodiment, game items in the gazing game screen may be identified, a library of game items for the identified game items and the target game combined, game items similar to, of the same class as, or used in conjunction with the identified game items may be found as potential game items, and game item recommendation information may be generated.

In one embodiment, referring to fig. 4, the generating game item recommendation information for a player according to the image recognition result may include the following steps S410 to S440:

step S410, determining the confidence of the potential game object and the potential game object according to the image recognition result.

The potential game items may include game items identified from the gaze game screen, and may also include game items similar to, of the same class as, or used in concert with the identified game items. The confidence of the potential game item may characterize a probability that the player gazes at or is interested in the potential game item. Illustratively, the image recognition result includes a probability of recognizing game item x1 in the gazing game screen as a, and game item x1 is taken as a potential game item, with confidence of game item x1 being a. If game item x2 similar to, of the same type as, or used in combination with game item x1 is also used as a potential game item, the confidence of game item x2 may be a.

In one embodiment, after the potential game item and the confidence level of the potential game item are determined based on the image recognition result, the confidence level of the potential game item may be modified. Illustratively, the potential game items may be ranked according to a first cause effect, with the confidence level being high to low, and the confidence level of the m potential game items ranked first may be reduced, e.g., the confidence level of the m potential game items may be reduced by decreasing proportions, respectively (e.g., m=3, the confidence level of the potential game item with the highest confidence level is reduced by 50%, the confidence level of the potential game item with the second highest confidence level is reduced by 30%, and the confidence level of the potential game item with the third highest confidence level is reduced by 20%). This makes the confidence distribution of potential game items more reasonable.

Step S420, determining the game item category to which the potential game item belongs as the potential game item category, and determining the quantized value of the potential game item set according to the category weight of the potential game item category and the confidence level of the potential game item.

Categorizing the potential game items can avoid overcomplicating the distribution of the potential game items. For example, multiple potential game items belonging to the same game item category may be normalized according to the game item category to which the potential game item belongs, e.g., when two potential game items belong to the same game item category (e.g., long sword), their respective weights x1, x2 are normalized to the category weight x1' of the game item category. In this exemplary embodiment, the game item category to which the potential game item belongs is referred to as a potential game item category, and the category weight of the potential game item category is obtained, where the category weight may be a weight preset for different game item categories in the target game, for example, the service importance degree of different game item categories in the target game may be considered, and a corresponding category weight is set for each game item category. And summarizing and quantifying the potential game items according to the category weight of the potential game item category and the confidence of the potential game items to obtain a quantified value of the potential game item set (namely, the set of all the potential game items). For example, the quantized values of the set of potential game items may be calculated by fitting the function f1 (x) =a·x1'+b·x2' + …. Where x1', x2' represent the class weights of the potential game item classes and a, b represent the confidence of the potential game items in the potential game item classes (if the confidence is modified, the modified confidence is employed). The quantified values of the set of potential game items may characterize the distribution characteristics of the game items that may be desired or of interest to the player.

In one embodiment, prior to determining the quantified value of the set of potential game items, the game item recommendation method may further comprise the steps of:

and correcting the category weight of the potential game item category according to the sequence of the player watching time corresponding to the potential game item category.

The player's watching time corresponding to the potential game item category may be the appearance time of the watching game picture where the potential game item in the potential game item category is located, and also be the collecting time of the eye movement data of the player watching the potential game item. The category weights may be modified based on first cause effects, and the more potential game items a player looks at, the greater the probability that the player will have a demand or interest in them. Therefore, the category weights of the potential game item categories can be corrected according to the sequence of the player watching time corresponding to the potential game item categories, so that the category weights of the potential game item categories with early watching time of the player are larger after correction. For example, the class weight of the potential game item class with the earliest player gaze time may be increased (e.g., by a factor greater than 1), and the class weights of the other potential game item classes may remain unchanged. Alternatively, a corresponding correction factor may be set for each potential game item category according to the chronological order of the player's gaze time corresponding to the potential game item category, such that the more advanced the potential game item category, the greater its corresponding correction factor. For example, if the number of potential game item categories is k, after sorting according to the order of the player's gaze time, correction coefficients greater than 1 are set for the first k/2 potential game item categories, correction coefficients less than 1 are set for the last k/2 potential game item categories, the sum of all k correction coefficients may be k, and the correction coefficients decrease in order. And multiplying the category weight of each potential game item category by the corresponding correction coefficient to realize correction of the category weight.

It should be appreciated that if the class weights for the potential game item classes are modified, the modified class weights may be used in calculating the quantized values for the potential game item sets.

In one embodiment, after determining the quantified value of the set of potential game items, the game item recommendation method may further include the steps of:

and if the player gazing time corresponding to the potential game object class is in the key time period in the game process, correcting the class weight of the potential game object class, and updating the quantized value of the potential game object set.

The key time period in the game process can be determined according to the game type or specific requirements. For example, a time period before and after a time point when a player applies a critical skill (e.g., a big sign) may be used as a critical time period, and if the unit time is 5 seconds, a time period before 5 seconds and a time period after 5 seconds (total time period of 10 seconds) when the critical skill is applied may be used as critical time periods.

Based on the near-cause effect, if the player's gaze time corresponding to the potential game item category is within the critical time period, the player may be considered to have a higher probability of requiring or being interested in the potential game item category, and the category weight of the potential game item category may be increased, for example, by a specific factor greater than 1 (the factor may be determined empirically or according to specific requirements) to achieve correction of the category weight.

In one embodiment, correction of category weights may be achieved by performing a secondary process on eye movement data for a critical period of time. Specifically, eye movement data in a key time period in a game process can be acquired, a fixation point in the key time period is determined according to the eye movement data, a plurality of fixation points with time intervals smaller than a first preset time and inter-coordinate distances smaller than a first preset distance are aggregated, a fixation area is determined according to the aggregated fixation points, and a fixation game picture is intercepted from game pictures in the key time period. And then, carrying out image recognition on the watched game screen, and if the recognized game object exists in one or more potential game object categories, carrying out the correction on the category weights of the one or more potential game object categories.

It should be appreciated that if the category weights for the potential game item categories are modified after the quantized values for the potential game item sets are determined, the quantized values for the potential game item sets may be updated based on the modified category weights.

The two ways of correcting the category weights of the potential game object categories based on the first cause effect and the near cause effect can enable the category weights to be more matched with the needs or interests of players, so that the accuracy of the quantized values of the potential game object sets is improved.

Step S430, determining whether the game item category to be matched is a recommended game item category based on the quantized values of the potential game item set and the quantized values of the multiple subcategories under the game item category to be matched in the target game.

The game item category to be matched can be any game item category in the target game, and can also be a specific game item category. By way of example, game items that may be desired or of interest to a player may be narrowed down to a certain extent in advance based on player characteristics, the type of character that the player uses, the game mode that the player selects, and so forth. If the type of character used by the player is a warrior, then the game item that the player may need or be interested in should be a warrior's game item, and the game item category to be matched may be any of the warrior's game items.

The game item category to be matched may include a plurality of subcategories, which may be a further subdivision of the game item category to be matched, such as finer item category divisions under the game item category to be matched to determine subcategories. In one embodiment, the game item categories to be matched may include multiple subcategories that are divided according to the price intervals of the game items, i.e., game items of different subcategories are within different price intervals.

The manner in which the quantized values of the subcategories are calculated may be referred to as the manner in which the quantized values of the set of potential game items are calculated. In one embodiment, the class weights of the subcategories may be weighted by the ratio of the number of game items of the subcategories in the class of game items to be matched to obtain a quantized value for the subcategory. The category weight of the subcategory can be obtained by taking the category weight of the game item category to be matched as a reference and carrying out certain adjustment, and the higher the price interval of the subcategory is, the larger the category weight of the subcategory can be.

Based on the quantized values of the potential game item set and the quantized values of the sub-categories of the game item to be matched in the target game, the similarity or difference condition of the quantized values of the potential game item set and the quantized values of the sub-categories can be judged, and if the quantized values of the potential game item set are similar to the quantized values of the sub-categories, the game item to be matched can be determined to be the game item to be recommended, namely, the player is judged to possibly have the requirement or interest on the game item to be matched.

In one embodiment, the game item categories to be matched include a plurality of sub-categories divided according to game item price intervals. The determining whether the game item category to be matched is the game item category to be recommended based on the quantized values of the potential game item set and the quantized values of the sub-categories under the game item category to be matched in the target game includes:

And calculating variances of the quantized values of the potential game item sets and the quantized values of the multiple subcategories, and if the variances are smaller than a preset variance threshold, determining the game item category to be matched as the game item category to be recommended.

For example, for a certain class of game items to be matched, the quantized values of its multiple subcategories are calculated as f2 (x), f3 (x), …, fn (x), respectively. Calculating variances of quantized values f1 (x) and f2 (x), f3 (x), … and fn (x) of the potential game item set, and if the variances are smaller than a preset variance threshold, considering that the player may have a need or interest in the game item category to be matched, and determining the game item category to be matched as the game item category to be recommended. The preset variance threshold may be determined empirically or based on specific needs. The quantized values of the potential game item sets are more similar to the quantized values of the subcategories in a way of calculating and comparing variances, so that the category of the game item to be recommended can be determined efficiently and accurately.

In one embodiment, each game item category in the target game may be sequentially used as a game item category to be matched, and whether the game item category is a game item category to be recommended is determined in the above manner. Therefore, more comprehensive game object categories to be recommended can be obtained.

Step S440, game item recommendation information is generated based on the game item category to be recommended.

All game items in the game item category to be recommended can be used as the game items to be recommended, certain screening can be performed to screen out the game items to be recommended, and the game items in the game item category to be recommended can be expanded or extended to use part of the game items outside the game item category to be recommended as the game items to be recommended, so that game item recommendation information is generated.

In one embodiment, the generating game item recommendation information based on the game item category to be recommended may include the steps of:

determining the game items to be recommended from the game item categories to be recommended based on the price information of each game item in the game item categories to be recommended and the characteristic information of the player;

a list of game items to be recommended is generated as game item recommendation information.

The characteristic information of the player may include demographic information (such as gender, age, region to which the player belongs), game characteristic information of the player (such as game preference determined according to historical game behavior of the player, game items already existing by the player), and the like. Based on the characteristic information of the player, a price interval preferred by the player can be determined, and in combination with the price information of each game item in the game item category to be recommended, the game item with the matched price can be determined as the game item to be recommended. Thereby further ensuring the matching of the game items to be recommended with the players.

After determining the game item to be recommended, the game item associated with the game item to be recommended can be added according to the predefined game item recommendation logic, and the game item can also be used as the game item to be recommended. Alternatively, game items whose price intervals or qualities match player characteristics may be screened from game items associated with game items to be recommended as game items to be recommended. For example, a game item to be recommended may be used in combination with a game item to be recommended, such as a long sword and a reinforcing stone, and if the long sword is determined to be a game item to be recommended, a reinforcing stone matching the player's preferred quality may be determined as a game item to be recommended.

Next, a list of game items to be recommended is generated as game item recommendation information.

With continued reference to fig. 2, in step S220, game item recommendation information is sent to the anchor 1202 associated with the target game in the live platform, such that the anchor 1202 generates live video for recommending the game item according to the game item recommendation information, the live video for presentation to the player.

The main broadcasting end 1202 associated with the target game may be a main broadcasting end 1202 corresponding to a live broadcasting room where a live broadcasting tag of the target game is preset, or a main broadcasting end 1202 corresponding to a live broadcasting room where data intercommunication of the target game is preset, etc.

The game server 1101 may send game item recommendation information to the live server 1201, which is forwarded by the live server 1201 to the host 1202. Alternatively, where the host side 1202 establishes a connection with the game server side 1101 (e.g., the host side 1202 logs in to the target game), the game server side 1101 may send game item recommendation information directly to the host side 1202.

After receiving the game item recommendation information, the presenter 1202 may perform live on-hold explanation of the game item to be recommended, thereby generating a live video. The live video may be sent to the game client 1102 where the player is located through the live server 1201 or the game server 1101, and finally presented to the player.

In one embodiment, the eye movement data in step S210 may be eye movement data in a game pair of the target game. The recommendation information of the game object can be determined based on the eye movement data in the game of one time, and finally the live video is generated. The game item recommendation method may further include the steps of:

in response to the game play ending, the live video is sent to the game client 1102 where the player is located to play the live video on the game client 1102.

The game server 1101 may send the live video to the game client 1102 after the game is played, so that the player may watch the live video immediately after the game is played, and the probability of matching the live video with the requirement or interest of the live video is high.

In order for the player to see the live video, an interface for playing the live video may be provided in the target game, such as a video player may be provided in the interface of the target game. When the game client 1102 where the player is located receives the live video, the interface may be automatically opened and the live video played. Or playing live video in the form of a window. In one embodiment, an interface of the target game may be implanted in the live interface, for example, when the game client 1102 where the player is located receives the live video, the player may be prompted to play the live video, if the player selects consent, the player jumps to the live interface to display the live video of the main game to explain the game item, and at the same time, a real-time screen of the target game is displayed in a window form in the live interface, so that the player can watch the live video and perform the operation in the game. Of course, the function of switching the game interface and the live interface can also be set, if the game interface is mainly displayed at present and the live video is implanted in the game interface to display the live video, after the player performs the switching operation (for example, a control for switching can be set, and the player clicks the control to perform switching), the main live interface can be switched and the game screen is implanted in the live interface.

In one embodiment, the game item recommendation method may further include the steps of:

receiving purchase behavior data of a game item purchase by a player under the condition of seeing a live video;

and updating the game resource data of the player according to the purchasing behavior data.

Wherein, when the player sees the live video, the player can purchase the game item directly in the live room (such as selecting the game item for purchase from the item bar of the live room), or can purchase the game item in the target game (such as purchasing the game item in the game store of the target game). For example, if a player makes a purchase request for a game item, the game item may be selected and purchased via an input device (e.g., a handle, etc.) of the headset. After the player purchases the game item, the live server 1201 or the game client 1102 may send the purchase behavior data of the player to the game server 1101, and the game server 1101 updates the game resource data of the player according to the purchase behavior data, where the game resource may include the game item, the game currency (or other game props for transaction). Illustratively, the game server 1101 may add its purchased game items to the player's item column based on the purchase behavior data (accordingly, the game client 1102 may render the game items in the player's item column) while deducting the corresponding game currency, thereby completing the update of the game resource data.

In one embodiment, if game server 1101 includes a plurality of servers with different functions, the game item recommendation method in the present exemplary embodiment may perform different steps or logic by different servers. For example, game server 1101 includes a game processing server, a game item transaction server, and the like. The method steps of fig. 2 may be performed by a game item transaction server, with the game processing server performing the processing of game plays.

Fig. 5 illustrates an exemplary flow of a game item recommendation method performed by the sponsor 1202, which may include the following steps S510 and S520:

step S510, game item recommendation information for players in the target game is acquired; game item recommendation information is generated from eye movement data of a player wearing the headset in a target game.

The game item recommendation information may be sent by the game server 1101 to the live broadcast server 1201, then sent by the live broadcast server 1201 to the anchor 1202, or directly sent by the game server 1101 to the anchor 1202.

In one embodiment, before obtaining game item recommendation information for a player in a target game, the host 1202 may also obtain characteristic information of the player (e.g., demographic data, player ID, game equipment data of the player, game play history data, etc.), information of the target game (e.g., ID of the target game, game type information, etc.) from the head-mounted device server 130, live server 1201, or game server 1101 so that the host can learn live-tie related information to facilitate the host's preparation in advance.

In one embodiment, in response to a player opening a target game or opening a game play in a target game, the head-mounted device server 130 or the game server 1101 sends characteristic information of the player and information of the target game to the live broadcast server 1201, the live broadcast server 1201 generates a live broadcast cargo opening instruction, sends the live broadcast cargo opening instruction, characteristic information of the player and information of the target game to the host 1202, and the host 1202 can learn live broadcast cargo of a game item to be opened and learn live broadcast cargo related information.

Step S520, generating a live video for recommending the game item according to the game item recommendation information, wherein the live video is used for showing to the player.

After acquiring the game item recommendation information, the anchor 1202 may perform live broadcast and on-hold explanation on the game item to be recommended, thereby generating a live broadcast video. The live video may be sent to the game client 1102 where the player is located through the live server 1201 or the game server 1101, and finally presented to the player.

a virtual scene is determined that matches the player.

Accordingly, the generating the live video for recommending the game item according to the game item recommendation information may include the following steps:

according to the game item recommendation information, rendering information of the game item to be recommended at a preset position in the virtual scene;

and combining the virtual scene with the live broadcast with-cargo picture formed by shooting the anchor, and generating a live broadcast video.

The virtual scene is a background in the live video, so that the immersion of a player watching the live video can be improved. The game item recommendation information may include information of the game item to be recommended, and the presenter 1202 may render the information of the game item to be recommended at a preset position in the virtual scene. The preset position is a position suitable for displaying live broadcast information, for example, the preset position can be a middle position of the virtual scene, or when the virtual scene comprises a shelf, the preset position can be a shelf position. The information of the game items to be recommended may be rendered into the virtual scene in any form, such as in the form of a game item list, a thumbnail of the game items on a shelf, and the like.

The camera of the presenter 1202 may take pictures of the presenter for live delivery (i.e., video explanation of the game item to be recommended). And combining the virtual scene with the live broadcast with the goods picture, for example, the live broadcast with the goods picture can be embedded into the virtual scene, so as to generate the live broadcast video.

Virtual scenes in the live video may be matched to the player, such as to conform to the player's preferences, so that the player is more interested in viewing the live video.

In one embodiment, the determining the virtual scene matching the player may include capturing:

acquiring characteristic information of a player;

and determining virtual scenes matched with the player and the target game according to the characteristic information of the player and the game type of the target game.

The characteristic information of the player may include demographic information, game characteristic information of the player, and the like. The Game types of the target Game may be MMORPG (Massively Multiplayer Online Role-play Game), MOBA (Multiplayer Online Battle Arena, multiplayer online tactical Game), FPS (First Person Shooting, first person shooter), etc., but may be more finely divided based on these Game types, such as may be added to a Game style, which may include the fairy style MMORPG, the mid-century style MMORPG. Modern war style MMORPGs and the like. The preference of the player can be estimated according to the characteristic information of the player, the virtual scene matched with the player is determined, the virtual scene matched with the style of the target game can be determined according to the game type of the target game, and finally the virtual scene matched with both aspects is determined. The player watches the live video while playing the game, the live video accords with the preference of the player, and the picture of the live video is matched with the game picture in harmony, so that the watching experience is better.

In one embodiment, referring to fig. 6, determining a virtual scene matching with the player and the target game according to the characteristic information of the player and the game type of the target game may include the following steps S610 to S650:

step S610, expected game item information is determined according to the game type of the target game.

The specific game type may correspond to specific game item information as intended game item information, which may include game item templates, game item styles, and the like. For example, the fairy-style MMORPG game type corresponds to game items conforming to the fairy character and the FPS game type corresponds to firearm game items, thereby determining the expected game item information. The game item identification (e.g., number or name) that may be involved may also be recorded in the desired game item information.

Step S620, scene preference information of the player is determined according to the characteristic information of the player.

For example, the player may be imaged according to information such as age, sex, historical game preference, etc. of the player, and thus the scene preference information thereof is estimated. The relationship between the characteristic information of the player and the scene preference information can be obtained by counting a large amount of player data, or can be obtained by learning by using a machine learning model. For example, a middle-aged MMORPG player may prefer a virtual scene of the style of the ancient swordsman. The player's scene preference information may include one or more scene style identification information that the player may prefer, or scene style categories that the player may prefer.

Step S630, matching the expected game item information with the game item information of the candidate virtual scene to obtain a first matching result.

The candidate virtual scenes are virtual scenes available in the live video. For example, the live platform may pre-configure a virtual scene library, where the virtual scenes are candidate virtual scenes. The game item information of the candidate virtual scene is used for characterizing which game items are suitable for the candidate virtual scene, and may include information such as category, style, color and the like of the game items suitable for the candidate virtual scene. Matching the expected game item information with the game item information of the candidate virtual scene can determine whether the game types of the candidate virtual scene and the target game are matched. The first matching result may be a matching degree of the expected game item information and the game item information of each candidate virtual scene, or may be a result of screening the candidate virtual scenes based on the matching degree of the game item information.

Step S640, matching the scene preference information of the player with the candidate virtual scenes to obtain a second matching result.

The scene preference information of the player may be matched with the style of the candidate virtual scene or features may be extracted from the candidate virtual scene to calculate a similarity with the scene preference information of the player. It is thus possible to determine whether the candidate virtual scene is adapted to the scene preference information of the player. The second matching result may be a matching degree of the scene preference information of the player and each candidate virtual scene, or may be a result of screening the candidate virtual scenes based on the matching degree of the scenes.

Step S650, determining a virtual scene matching the player and the target game from the candidate virtual scenes based on the first and second matching results.

And combining the first matching result and the second matching result, and evaluating the matching condition of the candidate virtual scene from two aspects of game item information matching and scene feature matching.

In one embodiment, the first matching result and the second matching result are two sets of screening candidate virtual scenes, and the first matching result and the second matching result are intersected to obtain virtual scenes matched with the player and the target game.

In one embodiment, the first matching result may include a first degree of matching of game item information of each candidate virtual scene with expected game item information; the second matching result may include a second degree of matching of each candidate virtual scene with scene preference information of the player; weighting the first matching degree and the second matching degree of each candidate virtual scene (the weight can be determined according to experience, specific requirements, business logic and the like, and if the scene style is more watched and matched, the second matching degree can be set with higher weight) so as to obtain the comprehensive matching degree of each candidate virtual scene. According to the comprehensive matching degree, the candidate virtual scene with the highest comprehensive matching degree can be selected, and the candidate virtual scenes with the higher comprehensive matching degree can be further screened. And finally determining the virtual scenes matched with the home and the target game.

In one embodiment, the anchor 1202 may add three-dimensional coordinate point data of a preset tile of a game frame available for rendering a target game in the virtual scene. When the virtual scene is rendered, the game patches can be rendered in a given area in the virtual scene according to the three-dimensional coordinate point data of the preset patches so as to present the game picture of the target game. Thus, the player or spectator terminal 1203 can see the game screen in the virtual scene.

In one embodiment, two view modes may be set for the virtual camera in the virtual scene, the first is a live global view mode, and in this view mode, the virtual camera may capture a game picture in the virtual scene and an area outside the game picture (may include an area of a background of the virtual scene, information of a game object to be recommended, etc.), so that the player or the spectator 1203 may see a relatively complete live picture in the live video. The second is a game view angle mode, in which the pose Of the virtual camera can be set according to the Field Of View (FOV) information Of the head-mounted device Of the player, so that the virtual camera only shoots a game picture in the virtual scene, and the game picture is the same as the view angle Of the picture that the player can actually see in the game, so that the player or the spectator 1203 can completely see the game picture in the live video, and the display effect is the same as the effect Of loading only the game picture.

The anchor side 1202 causes the player or viewer side 1203 to see live pictures of different effects by switching between the two viewing angles modes. For example, when the player is in the game play, the main player 1202 switches to the game view mode, and the live video including only the game screen is shot by the virtual camera, and the live video is sent to the live server 1201, further to the game server 1101, and then to the game client 1102, where the live video is played by the game client 1102, and the player can only see the game screen, and cannot see the live content outside the game screen, and the feel is the same as that of the normal game. When the game is completed, the main broadcasting end 1202 switches to a live broadcast global view mode, and shoots a live broadcast video with a complete picture through the virtual camera, wherein the live broadcast video comprises a game picture (such as a settlement picture of the game), a picture of a virtual scene, a live broadcast with goods picture and the like, and sends the live broadcast video to the live broadcast server 1201, further sends the live broadcast video to the game server 1101, and then sends the live broadcast video to the game client 1102, and the game client 1102 plays the live broadcast video, so that a player can see the live broadcast content outside the game picture and the game picture at the same time. By switching between the two view modes of the main cast 1202, the screen content of the game client 1102 can be flexibly controlled, so that the player can switch between a live view state and a fully-input state while playing.

For a player-controlled game client 1102, it may only receive and display live video in which the player can see a game scene, such as a live scene containing a game scene in a live global view mode, and only see a game scene in the live scene in a game view mode, thus eliminating the need to receive or display the game video.

In one embodiment, one or more of the above-described steps performed by the hosting side 1202 may be performed by an instance of a live engine (e.g., may be a phantom engine for generating a virtual live room) running on the hosting side 1202.

Exemplary embodiments of the present disclosure provide a game item recommendation device applied to a game server 1101 of a target game. Referring to fig. 7, the game item recommendation device 700 may include the following program modules:

an eye movement data processing module 710 configured to acquire eye movement data of a player wearing the head-mounted device in a target game, generate game item recommendation information for the player according to the eye movement data;

the game item recommendation information sending module 720 is configured to send the game item recommendation information to the anchor 1202 associated with the target game in the live platform, so that the anchor 1202 generates a live video for recommending the game item according to the game item recommendation information, and the live video is used for being shown to the player.

In one embodiment, the generating game item recommendation information for a player according to the eye movement data includes:

determining the gazing game picture of the player in the game process according to the eye movement data;

and carrying out image recognition on the gazing game picture, and generating game item recommendation information for the player according to the image recognition result.

In one embodiment, the determining, according to the eye movement data, a gaze of the player in the game process includes:

In one embodiment, the aggregating the gaze points per unit time includes:

In one embodiment, the generating game item recommendation information for the player according to the image recognition result includes:

Determining the potential game items and the confidence degrees of the potential game items according to the image recognition results;

determining the game item category to which the potential game item belongs as a potential game item category, and determining a quantization value of the potential game item set according to the category weight of the potential game item category and the confidence level of the potential game item;

determining whether the game item category to be matched is a game item category to be recommended based on the quantized values of the potential game item set and the quantized values of the plurality of subcategories under the game item category to be matched in the target game;

game item recommendation information is generated based on the game item category to be recommended.

In one embodiment, the eye movement data processing module 710 is further configured to:

before the quantized value of the potential game item set is determined, the category weights of the potential game item categories are corrected according to the sequence of the player gazing time corresponding to the potential game item categories.

after the quantized value of the potential game item set is determined, if the player's gaze time corresponding to the potential game item class is within a critical time period in the game process, the class weight of the potential game item class is modified, and the quantized value of the potential game item set is updated.

In one embodiment, the game item categories to be matched include a plurality of sub-categories divided according to game item price intervals; the determining whether the game item category to be matched is the game item category to be recommended based on the quantized values of the potential game item set and the quantized values of the sub-categories under the game item category to be matched in the target game includes:

In one embodiment, the generating game item recommendation information based on the game item category to be recommended includes:

In one embodiment, the eye movement data is eye movement data in a game play of a target game; the game item recommendation device 700 may further include a live video transmission module configured to:

In one embodiment, the game item recommendation device 700 may further include a purchase behavior data processing module configured to:

Exemplary embodiments of the present disclosure provide another game item recommendation device that is applied to a presenter 1202 associated with a target game in a live platform. Referring to fig. 8, the game item recommendation device 800 may include the following program modules:

a game item recommendation information acquisition module 810 configured to acquire game item recommendation information for a player in a target game; game item recommendation information is generated based on eye movement data of a player wearing the headset in a target game;

the live video generation module 820 is configured to generate live video for recommending game items according to game item recommendation information, the live video for presentation to a player.

In one embodiment, the game item recommendation device 800 may further include a virtual scene processing determination module configured to: determining a virtual scene that matches the player;

the generating a live video for recommending the game item according to the game item recommendation information includes:

In one embodiment, the determining the virtual scene matching the player includes:

acquiring characteristic information of a player;

In one embodiment, the determining the virtual scene matching the player and the target game according to the characteristic information of the player and the game type of the target game includes:

determining expected game item information according to the game type of the target game;

determining scene preference information of the player according to the characteristic information of the player;

matching the expected game item information with the game item information of the candidate virtual scene to obtain a first matching result;

Matching the scene preference information of the player with the candidate virtual scene to obtain a second matching result;

a virtual scene matching the player and the target game is determined from the candidate virtual scenes based on the first matching result and the second matching result.

In one implementation, live video generation module 820 is further configured to:

in a live broadcast global view mode, shooting a game picture in a virtual scene and an area except the game picture through a virtual camera arranged in the virtual scene of a live broadcast room, and generating a live broadcast picture for displaying to a player;

in the game view angle mode, the pose of the virtual camera is set according to the field of view information of the head-mounted device of the player, and only game pictures in the virtual scene are shot through the virtual camera, so that a live broadcast picture for displaying to the player is generated.

switching to a game viewing angle mode when the player is in a game play;

and when the game play is finished, switching to a live global view mode.

The specific details of each part in the above apparatus are already described in the method part embodiments, and the details not disclosed can refer to the embodiment content of the method part, so that the details are not repeated.

Exemplary embodiments of the present disclosure also provide a computer readable storage medium, which may be implemented in the form of a program product comprising program code for causing an electronic device to carry out the steps according to the various exemplary embodiments of the disclosure as described in the above section of the "exemplary method" when the program product is run on the electronic device. In an alternative embodiment, the program product may be implemented as a portable compact disc read only memory (CD-ROM) and comprises program code and may run on an electronic device, such as a personal computer. However, the program product of the present disclosure is not limited thereto, and in this document, a readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

The computer readable signal medium may include a data signal propagated in baseband or as part of a carrier wave with readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A readable signal medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Program code embodied on a readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Program code for carrying out operations of the present disclosure may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server. In the case of remote computing devices, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., connected via the Internet using an Internet service provider).

The exemplary embodiments of the present disclosure also provide an electronic device, such as a terminal. The electronic device may include a processor and a memory. The memory stores executable instructions of the processor, such as program code. The processor performs the method of the present exemplary embodiment by executing the executable instructions.

In one implementation, the electronic device may further include a display for displaying a graphical user interface in which a game screen, a live screen, etc. may be displayed.

An electronic device is illustrated in the form of a general purpose computing device with reference to fig. 9. It should be understood that the electronic device 900 shown in fig. 9 is merely an example and should not be construed as limiting the functionality and scope of use of embodiments of the present disclosure.

As shown in fig. 9, the electronic device 900 may include: processor 910, memory 920, bus 930, I/O (input/output) interface 940, and network adapter 950.

The memory 920 may include volatile memory, such as RAM 921, a cache unit 922, and may also include nonvolatile memory, such as ROM 923. Memory 920 may also include one or more program modules 924, such program modules 924 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment. For example, program modules 924 may include the modules in the apparatus described above.

The bus 930 is used to facilitate connections between the different components of the electronic device 900 and may include a data bus, an address bus, and a control bus.

The electronic device 900 may communicate with one or more external devices 1000 (e.g., keyboard, mouse, external controller, etc.) through an I/O interface 940.

The electronic device 900 may communicate with one or more networks through a network adapter 950, e.g., the network adapter 950 may provide a mobile communication solution such as 3G/4G/5G, or a wireless communication solution such as wireless local area network, bluetooth, near field communication, etc. The network adapter 950 may communicate with other modules of the electronic device 900 via the bus 930.

Although not shown in fig. 9, other hardware and/or software modules may also be provided in electronic device 900, including, but not limited to: displays, microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, data backup storage systems, and the like.

It should be noted that although in the above detailed description several modules or units of a device for action execution are mentioned, such a division is not mandatory. Indeed, the features and functionality of two or more modules or units described above may be embodied in one module or unit in accordance with exemplary embodiments of the present disclosure. Conversely, the features and functions of one module or unit described above may be further divided into a plurality of modules or units to be embodied.

Those skilled in the art will appreciate that the various aspects of the present disclosure may be implemented as a system, method, or program product. Accordingly, various aspects of the disclosure may be embodied in the following forms, namely: an entirely hardware embodiment, an entirely software embodiment (including firmware, micro-code, etc.) or an embodiment combining hardware and software aspects may be referred to herein as a "circuit," module "or" system. Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any adaptations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

It is to be understood that the present disclosure is not limited to the precise arrangements and instrumentalities shown in the drawings, and that various modifications and changes may be effected without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A game item recommending method is characterized by being applied to a game server of a target game; the method comprises the following steps:

acquiring eye movement data of a player wearing a head-mounted device in the target game, and generating game item recommendation information for the player according to the eye movement data;

and sending the game item recommendation information to a main broadcasting end associated with the target game in a live broadcasting platform, so that the main broadcasting end generates a live broadcasting video for recommending the game item according to the game item recommendation information, and the live broadcasting video is used for being displayed to the player.

2. The method of claim 1, wherein the generating game item recommendation information for the player from the eye movement data comprises:

and carrying out image recognition on the gazing game picture, and generating game item recommendation information for the player according to an image recognition result.

3. The method of claim 2, wherein said determining a gaze game screen of the player during the game based on the eye movement data comprises:

Determining a gaze point of the player during one or more unit times of the game based on the eye movement data;

4. A method according to claim 3, wherein said aggregating gaze points per unit time comprises:

5. The method of claim 2, wherein generating game item recommendation information for the player based on the image recognition result comprises:

determining a potential game item and a confidence level of the potential game item according to the image recognition result;

determining the game object class to which the potential game object belongs as a potential game object class, and determining a quantized value of a potential game object set according to the class weight of the potential game object class and the confidence level of the potential game object;

Determining whether the game item category to be matched is a game item category to be recommended based on the quantized values of the set of potential game items and the quantized values of the plurality of subcategories under the game item category to be matched in the target game;

and generating the game item recommendation information based on the game item category to be recommended.

6. The method of claim 5, wherein prior to determining the quantized values for the set of potential game items, the method further comprises:

and correcting the category weight of the potential game item category according to the sequence of the watching time of the player corresponding to the potential game item category.

7. The method of claim 5, wherein after determining the quantized values for the set of potential game items, the method further comprises:

and if the watching time of the player corresponding to the potential game object class is in a key time period in the game process, correcting the class weight of the potential game object class, and updating the quantized value of the potential game object set.

8. The method of claim 5, wherein the game item categories to be matched include the plurality of sub-categories divided according to game item price intervals;

The determining whether the game item category to be matched is a game item category to be recommended based on the quantized values of the set of potential game items and the quantized values of the plurality of subcategories under the game item category to be matched in the target game includes:

and calculating variances of the quantized values of the potential game item sets and the quantized values of the sub-categories, and if the variances are smaller than a preset variance threshold, determining the game item category to be matched as the game item category to be recommended.

9. The method of claim 5, wherein the generating the game item recommendation information based on the game item category to be recommended comprises:

determining a game item to be recommended from the game item category to be recommended based on price information of each game item in the game item category to be recommended and characteristic information of the player;

and generating the list of the game items to be recommended as the game item recommendation information.

10. The method of claim 1, wherein the eye movement data is eye movement data in a game play of the target game; the method further comprises the steps of:

and responding to the game play ending, and sending the live video to a game client where the player is located so as to play the live video on the game client.

11. The method according to claim 1, wherein the method further comprises:

receiving purchase behavior data of the player for purchasing the game object under the condition that the player sees the live video;

and updating game resource data of the player according to the purchase behavior data.

12. The game item recommending method is characterized by being applied to a main playing end associated with a target game in a live platform; the method comprises the following steps:

acquiring game item recommendation information for a player in the target game; the game item recommendation information is generated from eye movement data of the player wearing a head mounted device in the target game;

and generating a live video for recommending the game item according to the game item recommendation information, wherein the live video is used for being displayed to the player.

13. The method according to claim 12, wherein the method further comprises:

determining a virtual scene that matches the player;

the generating a live video for recommending the game item according to the game item recommendation information comprises the following steps:

rendering information of game items to be recommended at preset positions in the virtual scene according to the game item recommendation information;

And combining the virtual scene with a live broadcast with-cargo picture formed by shooting a host, and generating the live broadcast video.

14. The method of claim 13, wherein the determining a virtual scene that matches the player comprises:

acquiring characteristic information of the player;

15. The method of claim 14, wherein the determining a virtual scene matching the player and the target game based on the characteristic information of the player and the game type of the target game comprises:

matching the expected game item information with game item information of the candidate virtual scene to obtain a first matching result;

matching the scene preference information of the player with the candidate virtual scenes to obtain a second matching result;

based on the first and second matching results, a virtual scene that matches the player and the target game is determined from the candidate virtual scenes.

16. A game article recommending device is characterized by being applied to a game server side of a target game; the device comprises:

an eye movement data processing module configured to acquire eye movement data of a player wearing a head-mounted device in the target game, and generate game item recommendation information for the player according to the eye movement data;

the game item recommendation information sending module is configured to send the game item recommendation information to a main broadcasting end associated with the target game in a live broadcasting platform, so that the main broadcasting end generates a live broadcasting video for recommending the game item according to the game item recommendation information, and the live broadcasting video is used for being displayed to the player.

17. The game article recommending device is characterized by being applied to a main playing end which is associated with a target game in a live platform; the device comprises:

a game item recommendation information acquisition module configured to acquire game item recommendation information for a player in the target game; the game item recommendation information is generated from eye movement data of the player wearing a head mounted device in the target game;

and the live video generation module is configured to generate live video for recommending the game item according to the game item recommendation information, wherein the live video is used for being shown to the player.

18. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the method of any one of claims 1 to 15.

19. An electronic device, comprising:

a processor; and

a memory for storing executable instructions of the processor;

wherein the processor is configured to perform the method of any one of claims 1 to 15 via execution of the executable instructions.