Real-time facial animation control system and control method

Abstract

A real-time facial animation control system and a real-time facial animation control method are characterized in that information quality is guaranteed, efficiency of a storage area is improved, a storage unit is intelligently adapted, storage stability is guaranteed, various elements are introduced in a centralized manner, image information storage requirements and supply of the storage unit are accurately adapted, and loss of the storage unit is prevented. And the utilization rate of the storage area and the comprehensive management and control efficiency of external storage are improved, so that more efficient storage and management and control are achieved. According to the method, a centralized hierarchical measurement and self-adaptive distribution method is adopted, so that image information storage is more compliant and efficient, self-adaptive distribution of the image information and storage pages is achieved through information evaluation measurement, efficiency evaluation measurement, information attribute quantity and attached quantity operation and adaptation of all normal parameters of the image information, and the storage efficiency of the image information and the storage pages is improved. Compared with an existing image information storage method, the method has the advantages that different image information storage requirements can be met, and external storage maneuverability and high efficiency are improved.

Description

Technical Field

[0001] The invention belongs to the technical field of facial animation control, and in particular relates to a real-time facial animation control system and a control method. Background Art

[0002] The facial module can be used to create talking animated characters. Facial animation is an independent application tool created specifically for facial animation of 3D animated characters. Using the conversational writing format, it can determine the continuous position of the animated character's lips.

[0003] With the advancement of technology, the current real-time facial animation control can also expand its use, such as the existing technical solution with patent publication number "CN114972592A", which specifically includes a computer terminal with external memory connected to an image acquisition device. The image acquisition device is used to transmit the acquired facial image to the computer terminal, and the computer terminal is used to generate facial animation using the facial image as a key frame, and then store the facial image and its facial animation in the external memory. The facial image or its facial animation is the image information of the animation.

[0004] However, the current storage methods for animation image information still have many defects. For example, the image information is often not fully evaluated for quality before storage, which often results in a lot of inaccurate, incomplete or outdated image information in the external memory, which is not conducive to the quality and applicability of image information analysis. In addition, the allocation of storage areas is often based on a fixed method, lacking a more accurate grasp and flexible adjustment capabilities of image information attributes and browsing modes. Summary of the invention

[0005] In order to solve the defects in the prior art, the present invention proposes a real-time facial animation control system and control method, which ensures information quality, improves storage area efficiency, intelligently adapts storage units, ensures storage stability, centrally introduces multiple types of elements, accurately adapts image information storage requirements and storage unit supply, prevents storage unit wear, improves storage area utilization and overall management efficiency of external storage, so as to achieve more efficient storage and management. The present invention uses a centralized hierarchical measurement and adaptive distribution method to make image information storage more compliant and efficient. Through the calculation and adaptation of the information evaluation quantity, performance evaluation quantity, information attribute quantity and all normal parameters of the image information, the adaptive distribution of image information and storage pages is achieved, and the overall performance and utilization rate of external storage are improved. Compared with the current image information storage method, it can meet different image information storage requirements and improve the mobility and efficiency of external storage.

[0006] The present invention utilizes the following technical solutions.

[0007] A real-time facial animation control method, comprising:

[0008] The image acquisition device transmits the acquired face image to the computer terminal, and the computer terminal confirms the face image to obtain its facial animation, and then stores the face image and its facial animation in an external memory. The face image or its facial animation is the image information of the animation;

[0009] The method for storing a face image and its facial animation in an external memory comprises:

[0010] Step 1, collecting the image information of the animation to be stored;

[0011] Step 2, analyzing the information quality of any image information in the image information of the animation to be stored, and obtaining the information evaluation value of the image information; comparing the information evaluation value with its quality threshold value, if it is higher than its quality threshold value, it means that the quality of the image information does not meet the conditions, and a re-execution mode is formed; otherwise, if it is not higher than the quality threshold value, it means that the quality of the image information meets the conditions;

[0012] Step 3, evaluating the performance of the external memory storage area and performing improvement treatment, obtaining the performance evaluation value of the external memory storage area, and comparing it with the corresponding pre-defined critical quantity to obtain the corresponding performance treatment mode;

[0013] Step 4, according to the image information of the animation, the storage unit required for the information is analyzed to obtain the information attribute quantity; the numerical range of the low, medium and high attributes of the image information is set, and the information attribute quantity is adapted to the numerical range of the low, medium and high attributes to obtain the attributes of the corresponding image information, and the corresponding storage unit is adapted according to the attributes of the image information;

[0014] Step 5, evaluating the stability of each storage area in the external memory, and obtaining the overall normal parameters of a random page in the storage unit of the storage area;

[0015] Step 6, performs additional quantity calculation on the overall normal parameters and performance evaluation quantities to obtain the overall level quantity of the page; performs additional quantity calculation on the information evaluation quantity and information attribute quantity of the image information to be stored in the corresponding storage unit, and distributes the image information after obtaining the overall level quantity of the information.

[0016] Further, in step 4, the attributes include low information attributes, medium information attributes, and high information attributes;

[0017] If it is a low information attribute, the image information is prepared to be stored in the storage unit of the SRAM area of ​​the external memory; if it is a medium information attribute, the image information is prepared to be stored in the DDRAM area of ​​the external memory and then migrated to the storage unit of the DRAM area after a defined period of time; if it is a high information attribute, the image information is prepared to be stored in the storage unit of the DDRAM area of ​​the external memory.

[0018] Furthermore, in step 6, the method for distributing image information is executed, including: distributing the image information with the largest overall information level to the storage page with the largest overall information level of its corresponding storage unit; removing the distributed image information and pages from the queue to be distributed, and then distributing the next image information in accordance with the method of distributing the image information with the largest overall information level to the storage page with the largest overall information level of the corresponding storage unit, until all image information in the animation image information is distributed or there is no appropriate storage page.

[0019] Furthermore, the method for analyzing the information quality of any image information in the image information of the animation to be stored includes:

[0020] Obtain information receiving information corresponding to the image information of the animation, the information receiving information includes the information forming capacity and the information forming temporary storage capacity of the image information; confirm the information piece information of any image information in the image information of the animation, the information piece information includes the information transmission capacity and the information main storage capacity of the image information.

[0021] Furthermore, the information formation capacity is the capacity when the image information is formed, the information formation temporary storage capacity is the capacity occupied by the temporary storage when the graphic information is formed and transmitted to the computer terminal, the information transmission capacity is the capacity when the image information is taken out of the temporary storage and is to be transmitted to the main memory of the computer terminal, and the information main memory capacity is the capacity occupied by the image information when it is transmitted to the main memory.

[0022] Furthermore, the method for analyzing the information quality of any image information in the image information of the animation to be stored further includes:

[0023] Processing is performed on any image information in the image information of the animation corresponding to the image information of the animation, and the hash value of the random image information is defined as IG1 by using the hash method; the information formation capacity and the information formation temporary storage capacity of the random image information are respectively subtracted from the information transmission capacity and the information main memory capacity, so as to obtain the information capacity spacing amount and the area spacing amount respectively, which are defined as IG2 and IG3 respectively; the hash value of the image information, the information capacity spacing amount and the area spacing amount are calculated, and the equation IG=IG1*i1+IG2*i2+IG3*i3 is used to obtain the information evaluation amount IG; here, i1, i2, i3 respectively represent the criticality coefficients corresponding to the pre-set hash value, the information capacity spacing amount and the area spacing amount.

[0024] Further, the method of evaluating the performance of the external memory storage area and performing improved processing includes:

[0025] Confirm the performance parameters of the external memory storage area, perform spacing calculation on any parameter in the performance parameters and its corresponding base parameter, and obtain the performance spacing corresponding to the parameter, which is defined as H k , k represents the serial number of the parameter corresponding to the performance spacing quantity;

[0026] The performance gap of all performance parameters in the external memory area is calculated using the equation Get performance evaluation measure IH1; here, A pre-set key quantity representing the performance gap quantity corresponding to the sequence number k;

[0027] Set a corresponding pre-defined critical quantity for the performance evaluation, and compare the performance evaluation quantity with its corresponding pre-defined critical quantity to obtain a corresponding performance handling mode, that is, if the performance evaluation quantity is lower than its corresponding pre-defined critical quantity, a corresponding coping mode is formed.

[0028] Furthermore, the method of parsing the storage unit required for the information according to the image information of the animation includes:

[0029] The browsing frequency of the same image information as any image information in the image information of the animation is defined as Lu, u represents the serial number of the same image information, and the operation is performed based on the browsing frequency of any image information and the average browsing frequency of the same image information, using the equation Obtain the browsing frequency measure LH of the same image information; here, Lu represents the browsing frequency of the image information corresponding to the sequence number u, represents the average browsing frequency of image information of the same category, and uδ represents the number The corresponding key quantity;

[0030] Construct a task weight detection message table, send any image information of the animation image information into the task weight detection message table to obtain, send different detection methods to obtain the task weight of any image information and define it as Gs, s represents the serial number of the detection method contained in the task weight detection message table, and perform calculations on the task weight of any image information using different detection methods, using the equation GS = ∑ s＝1 (sGS*sι), obtain the weight level GS; here, sGS and sι represent the task weight corresponding to the detection method with sequence number s and the corresponding key quantity set in advance;

[0031] Get the ease of acquisition information of the animated image information, and process the ease of acquisition information, that is, use the equation QH = ∑ q＝1 (qLH*qν), obtain the abstract value QH;

[0032] The browsing frequency metric LH, the weight level GS, and the abstract value QH are quantified and the equation is used. The information attribute quantity TY is obtained; here, y1, y2, and y3 represent the pre-set key effect coefficients corresponding to the browsing frequency measurement, the weight level, and the abstract cost quantity, respectively.

[0033] Furthermore, the method for evaluating the stability of each storage area in the external memory includes:

[0034] Extract the residual storage capacity of each storage area in the external memory, and define the storage area whose residual storage capacity is higher than the storage capacity threshold value set in advance as the available storage area; set the maximum access and clearing frequency CCzg of the available storage area; obtain the access and clearing frequency CCxz of different pages corresponding to the storage units of each available storage area; use the equation CCt=CCzg-CCxz to subtract the access and clearing frequency from the maximum access and clearing frequency to obtain the residual access and clearing frequency CCt;

[0035] The quotient obtained by dividing the frequency of virus attacks on each page by the frequency of page access and clearing is used as the page stability parameter HE; the quotient obtained by dividing the frequency of page failure by the frequency of page access and clearing is used to obtain the failure anomaly probability HT; the residual access and clearing frequency, stability coefficient, and failure anomaly probability are processed, and the equation HC = (CCt*e1+HE*e2) / (HT*e3+1) is used to obtain the overall normal parameter HC of the page; here, e1, e2, and e3 respectively represent the pre-set key quantities corresponding to the residual access and clearing frequency, stability coefficient, and failure anomaly probability.

[0036] Furthermore, when forming the information re-execution mode, it includes:

[0037] Step 2-1, collect new image information again, then use the median filter method to filter the new image information and then perform step 2 to obtain a new information evaluation measure;

[0038] Step 2-2, set the repetition frequency, and obtain a new information evaluation quantity for each repetition;

[0039] Step 2-3, compare the maximum new information evaluation value with the quality threshold value set in advance. If it is higher than the quality threshold value, it means that the quality of the image information does not meet the requirements, and a re-selection process is formed; otherwise, if it is not higher than the quality threshold value, it means that the quality of the image information meets the requirements;

[0040] Step 2-4, when forming the re-taking fluctuation process, the new information evaluation quantity and the initial information evaluation quantity are subjected to the interval quantity operation to obtain the re-taking evaluation interval quantity; the re-taking evaluation interval quantity is subjected to the standard deviation operation to obtain the evaluation fluctuation value; the numerical interval of the evaluation frequency cluster is set, and the evaluation fluctuation value is adapted to the numerical interval of its evaluation frequency cluster to obtain the corresponding evaluation frequency; the evaluation frequency is used to increase the repetition frequency;

[0041] Step 2-5, execute step 2-3 again. If the maximum new information evaluation value is still higher than its quality threshold, remove the image information and form a discarded information entry with all discarded image information. If the new information evaluation value is not higher than its quality threshold, it means that the quality of the image information meets the conditions.

[0042] A real-time facial animation control system, comprising:

[0043] A computer terminal with an external memory connected to the image acquisition device, the image acquisition device is used to transmit the acquired face image to the computer terminal, the computer terminal is used to generate a facial animation using the face image as a key frame, and then the face image and the facial animation are stored in the external memory, the face image or the facial animation is the image information of the animation;

[0044] The modules running on the computer terminal include:

[0045] A receiving module, which is used to receive the image information of the animation to be stored;

[0046] The analysis module is used to analyze the information quality of any image information in the image information of the animation to be stored, and obtain the information evaluation value of the image information; compare the information evaluation value with its quality threshold value, if it is higher than its quality threshold value, it means that the quality of the image information does not meet the conditions, and the information re-execution mode is formed; otherwise, if it is not higher than the quality threshold value, it means that the quality of the image information meets the conditions;

[0047] An evaluation module is used to evaluate the performance of the external memory storage area and perform improvement treatment, obtain the performance evaluation value of the external memory storage area, and compare it with the corresponding pre-defined critical quantity to obtain the corresponding performance treatment mode;

[0048] A setting module is used to analyze the storage unit required for the information according to the image information of the animation to obtain the information attribute quantity; set the numerical range of the low, medium and high attributes of the image information, and adapt the information attribute quantity to the numerical range of the low, medium and high attributes to obtain the attributes of the corresponding image information, and adapt the corresponding storage unit according to the attributes of the image information;

[0049] A stability module is used to evaluate the stability of each storage area in the external memory and obtain the overall normal parameters of any page in the storage unit of the storage area;

[0050] The distribution module is used to perform additional quantity operations on the overall normal parameters and performance evaluation quantities to obtain the overall level quantity of the page; perform additional quantity operations on the information evaluation quantities and information attribute quantities of the image information to be stored in the corresponding storage unit, and distribute the image information after obtaining the overall level quantity of the information.

[0051] The beneficial effects of the present invention are that, compared with the prior art, the technical effects of the present invention include:

[0052] By ensuring information quality, improving storage area performance, intelligently adapting storage units, and ensuring storage stability, multiple types of elements are centrally introduced to accurately adapt image information storage requirements and storage unit supply, prevent storage unit wear, improve storage area utilization and overall management efficiency of external storage, so as to achieve more efficient storage and management. The present invention uses a centralized hierarchical measurement and adaptive distribution method to make image information storage more compliant and efficient. Through the calculation and adaptation of the information evaluation quantity, performance evaluation quantity, information attribute quantity and all normal parameters of the image information, the adaptive distribution of image information and storage pages is achieved, and the overall performance and utilization rate of external storage are improved. Compared with the current image information storage method, it can meet different image information storage requirements and improve the mobility and efficiency of external storage. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 is a flow chart of the real-time facial animation control method of the present invention;

[0054] Figure 2 It is a partial structural diagram of the real-time facial animation control system described in the present invention. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present invention clearer, the following will be combined with the drawings in the embodiments of the present invention to clearly and completely express the technical solution of the present invention. The embodiments expressed in this application are only some embodiments of the present invention, not all embodiments. According to the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the protection scope of the present invention.

[0056] like Figure 1 As shown, a real-time facial animation control method of the present invention comprises:

[0057] The image acquisition device transmits the acquired face image to the computer terminal, and the computer terminal uses the face image as a key frame to generate a facial animation, and then stores the face image and the facial animation in an external memory. The face image or the facial animation is the image information of the animation;

[0058] The method for storing a face image and its facial animation in an external memory comprises:

[0059] Step 1, collect the animated image information to be stored; here, the animated image information is formed by multiple face images or their facial animations; the face images or their facial animations represent different attributes of the image information.

[0060] Step 2, analyzing the information quality of any image information in the image information of the animation to be stored, and obtaining the information evaluation value of the image information; comparing the information evaluation value with its quality threshold value, if it is higher than its quality threshold value, it means that the quality of the image information does not meet the conditions, and the information re-execution mode is formed; otherwise, if it is not higher than the quality threshold value, it means that the quality of the image information meets the conditions, and it can continue to be stored;

[0061] Step 3, evaluating the performance of the external memory storage area and performing improvement treatment, obtaining the performance evaluation value of the external memory storage area, and comparing it with the corresponding pre-defined critical quantity to obtain the corresponding performance treatment mode;

[0062] Step 4, according to the image information of the animation, the storage unit required for the information is analyzed to obtain the information attribute quantity; the numerical range of the low, medium and high attributes of the image information is set, and the information attribute quantity is adapted to the numerical range of the low, medium and high attributes to obtain the attributes of the corresponding image information, and the corresponding storage unit is adapted according to the attributes of the image information;

[0063] In a preferred but non-limiting implementation method of the present invention, in step 4, the attributes include low information attributes, medium information attributes, and high information attributes;

[0064] If it is a low information attribute, the image information is prepared to be stored in the storage unit of the SRAM area of ​​the external memory, which is suitable for efficient browsing; if it is a medium information attribute, the image information is prepared to be stored in the DDRAM area of ​​the external memory and then migrated to the storage unit of the DRAM area after a defined period of time; if it is a high information attribute, the image information is prepared to be stored in the storage unit of the DDRAM area of ​​the external memory.

[0065] Step 5, evaluating the stability of each storage area in the external memory, and obtaining the overall normal parameters of a random page in the storage unit of the storage area;

[0066] Step 6, performs additional quantity calculation on the overall normal parameters and performance evaluation quantities to obtain the overall level quantity of the page; performs additional quantity calculation on the information evaluation quantity and information attribute quantity of the image information to be stored in the corresponding storage unit, and distributes the image information after obtaining the overall level quantity of the information.

[0067] In the preferred but non-restrictive implementation method of the present invention, in step 6, the method of distributing image information includes: distributing the image information with the largest overall information level to the storage page with the largest overall information level of the page of its corresponding storage unit; removing the distributed image information and page from the queue to be distributed, and then distributing the next image information in accordance with the method of distributing the image information with the largest overall information level to the storage page with the largest overall information level of the page of the corresponding storage unit, until all the image information in the image information of the animation is distributed or there is no appropriate storage page. Distribution means storage, and the queue to be distributed is a queue set in advance with the image information to be distributed and the page to be distributed. The page is also a storage page, which means a page storing image information.

[0068] By analyzing and evaluating the image information and external memory of the animation, adaptive storage and management of the image information is achieved, which improves the storage efficiency and stability of the image information. According to the quality and attributes of the image information, appropriate storage units and storage pages can be selected to ensure the storage and browsing efficiency of the data, and improved processing can be performed based on the performance and stability of the storage area to provide better storage quality. The present invention can be used in various animation image information environments to assist in external memory management and storage of large image information, and improve the efficiency of image information processing and analysis.

[0069] In a preferred but non-limiting implementation method of the present invention, the method of analyzing the information quality of any image information in the image information of the animation to be stored includes:

[0070] Obtain information receiving information corresponding to the image information of the animation, the information receiving information includes the information forming capacity and the information forming temporary storage capacity of the image information; confirm the information piece information of any image information in the image information of the animation, the information piece information includes the information transmission capacity and the information main storage capacity of the image information.

[0071] In the preferred but non-restrictive implementation method of the present invention, the information formation capacity is the capacity when the image information is formed. If the image information is a face image, then the capacity when the image information is formed is the capacity of the face image when the computer terminal receives the face image transmitted by the image acquisition device. If the image information is a facial animation of the face image, the capacity when the image information is formed is the capacity when the facial animation of the face image is confirmed. The information formation temporary storage capacity is the capacity occupied by the temporary storage of the computer terminal after the graphic information is formed. The information transmission capacity is the capacity when the image information is taken out of the temporary storage and is to be transmitted to the main memory of the computer terminal. The information main memory capacity is the capacity occupied by the image information when it is transmitted to the main memory. The image information is finally transmitted to the external memory via the image information sent to the main memory.

[0072] In a preferred but non-limiting implementation method of the present invention, the method for analyzing the information quality of any image information in the image information of the animation to be stored further includes:

[0073] Processing is performed on any image information in the image information of the animation corresponding to the image information of the animation, and the hash value of the random image information is defined as IG1 by using the hash method; the information formation capacity and the information formation temporary storage capacity of the random image information are respectively subtracted from the information transmission capacity and the information main memory capacity, so as to obtain the information capacity spacing amount and the area spacing amount respectively, which are defined as IG2 and IG3 respectively; the hash value of the image information, the information capacity spacing amount and the area spacing amount are calculated, and the equation IG=IG1*i1+IG2*i2+IG3*i3 is used to obtain the information evaluation amount IG; here, i1, i2, i3 respectively represent the criticality coefficients corresponding to the pre-set hash value, the information capacity spacing amount and the area spacing amount.

[0074] The hash method is used to calculate the hash value of the image information, perform operations on the information capacity spacing and the area spacing, and combine the criticality coefficient to obtain the final information evaluation value, which helps the external storage to grasp the actual quality status of each image information in the animated image information, and efficiently identify the situation where the information quality does not meet the requirements, providing a more accurate basis for the subsequent image information storage.

[0075] In a preferred but non-limiting implementation of the present invention, a method for evaluating the performance of an external memory storage area and performing improved processing includes:

[0076] Confirm the performance parameters of the external memory storage area, perform spacing calculation on any parameter in the performance parameters and its corresponding base parameter, and obtain the performance spacing corresponding to the parameter, which is defined as H k , k represents the serial number of the parameter corresponding to the performance spacing; here, the performance parameters include the access rate, access frequency, flow, mean obstacle time interval, time, transmission volume, information transmission rate, error frequency, and residual storage capacity of the external storage area;

[0077] The access rate is the rate at which image information is retrieved or stored in the external memory. The access frequency represents the frequency of access (that is, retrieving and storing image information) that can be executed in 1 second. The throughput is the capacity of image information retrieved or stored in 1 second. The mean failure time interval is the length of time that the external memory can work normally between two failures (failure is the situation that the external memory cannot normally perform access to image information). The residual storage capacity represents the capacity of the unused area in the external memory storage area. The usage time represents the length of time used to retrieve or store image information. The transmission volume represents the capacity of information that can be transmitted in the set time. The information transmission rate represents the rate at which image information is migrated in the external memory or between the main memory. The error frequency represents the frequency of incorrect image information storage in the external memory (an error is the situation of incorrect image information).

[0078] The performance gap of all performance parameters in the external memory area is calculated using the equation Get performance evaluation measure IH1; here, A pre-set key quantity representing the performance gap quantity corresponding to the sequence number k;

[0079] Set a pre-defined critical value corresponding to the performance evaluation, and compare the performance evaluation value with its corresponding pre-defined critical value to obtain the corresponding performance handling mode, that is, if the performance evaluation value is lower than its corresponding pre-defined critical value, a corresponding coping mode is formed; just like the performance handling mode corresponding to the access rate, access frequency, flow, mean obstacle time interval, time, transmission volume, information transmission rate, and error frequency, if a handling mode corresponding to the access rate or information transmission rate is formed, the storage unit is improved by methods such as cleaning up the fragments of the storage unit; if a handling mode corresponding to the flow is formed, the access operation requirements are decomposed through methods such as the average shortest response time to improve its overall flow; if a handling mode corresponding to the error frequency or the mean obstacle time interval is formed, adjustments are made such as distributing the image information data on multiple pages, and using the CRC error detection method to achieve improved information duplication and confidence; if a critical value corresponding to the residual storage capacity is formed, additional external memory is required.

[0080] By evaluating the performance of external memory storage areas and improving processing methods, real-time browsing and transmission of image information can be guaranteed, and the response speed and stability of external memory can be improved, thereby improving its overall storage efficiency and image information processing performance.

[0081] In a preferred but non-limiting implementation method of the present invention, a method for parsing the storage unit of the information of the animation according to the image information of the animation includes:

[0082] The browsing frequency of the same image information as any image information in the image information of the animation is defined as Lu, u represents the serial number of the same image information, and the operation is performed based on the browsing frequency of any image information and the average browsing frequency of the same image information, using the equation Obtain the browsing frequency measure LH of the same image information; here, Lu represents the browsing frequency of the image information corresponding to the sequence number u, represents the average browsing frequency of image information of the same category, and uδ represents the number The corresponding key quantity; the access frequency is the frequency of being accessed and used when the image information is generated and sent to the main memory.

[0083] Construct a task weight detection message table, send any image information of the animation image information into the task weight detection message table to obtain, send different detection methods to obtain the task weight of any image information and define it as Gs, s represents the serial number of the detection method contained in the task weight detection message table, and perform calculations on the task weight of any image information using different detection methods, using the equation GS = ∑ s＝1 (sGS*sι), obtain the weight level GS; here, sGS, sι respectively represent the task weight corresponding to the detection method with serial number s and the corresponding key quantity set in advance, and the task weight detection message table includes different detection methods formed by tools such as Google Analytics tool method, Mixpanel tool method, Apache Hadoop tool method, etc. to perform overall evaluation task weight on the input image information;

[0084] Get the ease of acquisition information of the animated image information, and process the ease of acquisition information, that is, use the equation QH = ∑ q＝1 (qLH*qν), obtain the extraction cost QH; here, the acquisition ease message includes the hardware cost of image information acquisition (just like the software and hardware cost of image acquisition equipment and computer terminals when forming image information), labor cost (just like the labor cost when forming image information) and time cost (the time taken to form image information), and these costs are given a sequence number q, qLH, qν respectively represent the acquisition cost with sequence number q in the acquisition ease message and its corresponding key quantity set in advance;

[0085] The browsing frequency metric LH, the weight level GS, and the abstract value QH are quantified and the equation is used. The information attribute quantity TY is obtained; here, y1, y2, and y3 represent the pre-set key effect coefficients corresponding to the browsing frequency measurement, the weight level, and the abstract cost quantity, respectively.

[0086] By calculating parameters such as browsing frequency, task volume and acquisition cost, and obtaining information attribute quantity according to the criticality coefficient, the low, medium and high information attributes of the image information are determined, and the appropriate storage unit is selected for storage. According to the different attributes of the image information, the image information can be stored in storage units with different speeds and costs to improve the storage efficiency of the image information and reduce the storage cost. By improving the selection of storage units, it can well meet the browsing requirements of the image information and improve the external memory performance and the flexibility of external memory use.

[0087] In a preferred but non-limiting implementation method of the present invention, the method for evaluating the stability of each storage area in the external memory includes:

[0088] Extract the residual storage capacity of each storage area in the external memory, and define the storage area whose residual storage capacity is higher than the pre-set storage capacity threshold as the available storage area; set the maximum access and clearing frequency CCzg of the available storage area; obtain the access and clearing frequency CCxz of different pages corresponding to the storage units of each available storage area; use the equation CCt=CCzg-CCxz to subtract the access and clearing frequency from the maximum access and clearing frequency to obtain the residual access and clearing frequency CCt; access and clearing means retrieving information, storing information and clearing pages.

[0089] The quotient obtained by dividing the frequency of each page being attacked by viruses by the frequency of page access and clearing is used as the page stability parameter HE; the quotient obtained by dividing the frequency of the page being blocked due to abnormalities (that is, unable to retrieve and store information normally) by the frequency of page access and clearing is obtained to obtain the obstacle abnormality probability HT; the residual access and clearing frequency, stability coefficient, and obstacle abnormality probability are processed, and the equation HC = (CCt*e1+HE*e2) / (HT*e3+1) is used to obtain the overall normal parameter HC of the page; here, e1, e2, and e3 respectively represent the pre-set key quantities corresponding to the residual access and clearing frequency, stability coefficient, and obstacle abnormality probability.

[0090] Evaluating the stability of each storage area ensures the logic and operability of the stability evaluation method, and helps control and manage the stability and normal status of external storage.

[0091] After step 6, a refreshed association table is also included to register the pointer of the newly distributed image information and the used storage area.

[0092] Ensuring the accuracy and completeness of data can help controllers track and monitor the usage of external storage.

[0093] In a preferred but non-limiting implementation method of the present invention, when forming an information re-execution mode, it includes:

[0094] Step 2-1, collect new image information again, then use the median filter method to filter the new image information and then perform step 2 to obtain a new information evaluation measure;

[0095] Step 2-2, set the repetition frequency, and obtain a new information evaluation quantity for each repetition;

[0096] Step 2-3, compare the maximum new information evaluation value with the quality threshold value set in advance. If it is higher than the quality threshold value, it means that the quality of the image information does not meet the requirements, and a re-selection process is formed; otherwise, if it is not higher than the quality threshold value, it means that the quality of the image information meets the requirements;

[0097] Step 2-4, when forming the re-taking fluctuation process, perform spacing operation on the new information evaluation quantity and the initial information evaluation quantity to obtain the re-taking evaluation spacing quantity; perform standard deviation operation on the re-taking evaluation spacing quantity to obtain the evaluation fluctuation value; set the numerical interval of the evaluation frequency cluster, and adapt the evaluation fluctuation value with the numerical interval of its evaluation frequency cluster to obtain the corresponding evaluation frequency; the evaluation frequency is used to increase the repetition frequency; adaptation is to find the same value.

[0098] Step 2-5, execute step 2-3 again. If the maximum new information evaluation value is still higher than its quality threshold, the image information is removed to indicate that the information quality still does not meet the requirements. The image information is removed and all discarded image information is formed into a discarded information entry. If the new information evaluation value is not higher than its quality threshold, it indicates that the quality of the image information meets the requirements.

[0099] By repeatedly collecting image information and evaluating the information quality over multiple rounds, qualified high-quality image information is selected, the deviation and volatility of the image information is reduced, the accuracy and reliability of the image information is improved, the subsequent image information analysis and accuracy are guaranteed, and the efficiency and reliability of image information processing are improved.

[0100] like Figure 2 As shown, a real-time facial animation control system according to the present invention comprises:

[0101] A computer terminal with an external memory connected to the image acquisition device, the image acquisition device is used to transmit the acquired face image to the computer terminal, the computer terminal is used to generate a facial animation using the face image as a key frame, and then the face image and the facial animation are stored in the external memory, the face image or the facial animation is the image information of the animation;

[0102] The modules running on the computer terminal include:

[0103] A receiving module, which is used to receive the image information of the animation to be stored;

[0104] The analysis module is used to analyze the information quality of any image information in the image information of the animation to be stored, and obtain the information evaluation value of the image information; compare the information evaluation value with its quality threshold value, if it is higher than its quality threshold value, it means that the quality of the image information does not meet the conditions, and the information re-execution mode is formed; otherwise, if it is not higher than the quality threshold value, it means that the quality of the image information meets the conditions;

[0105] An evaluation module is used to evaluate the performance of the external memory storage area and perform improvement treatment, obtain the performance evaluation value of the external memory storage area, and compare it with the corresponding pre-defined critical quantity to obtain the corresponding performance treatment mode;

[0106] A setting module is used to analyze the storage unit required for the information according to the image information of the animation to obtain the information attribute quantity; set the numerical range of the low, medium and high attributes of the image information, and adapt the information attribute quantity to the numerical range of the low, medium and high attributes to obtain the attributes of the corresponding image information, and adapt the corresponding storage unit according to the attributes of the image information;

[0107] A stability module is used to evaluate the stability of each storage area in the external memory and obtain the overall normal parameters of any page in the storage unit of the storage area;

[0108] The distribution module is used to perform additional quantity operations on the overall normal parameters and performance evaluation quantities to obtain the overall level quantity of the page; perform additional quantity operations on the information evaluation quantities and information attribute quantities of the image information to be stored in the corresponding storage unit, and distribute the image information after obtaining the overall level quantity of the information.

[0109] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not deviate from the spirit and scope of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A real-time facial animation control method, characterized in that: include: The image acquisition device transmits the acquired face image to the computer terminal, and the computer terminal uses the face image as a key frame to generate a facial animation, and then stores the face image and the facial animation in an external memory. The face image or the facial animation is the image information of the animation; The method for storing a face image and its facial animation in an external memory comprises: Step 1, collecting the image information of the animation to be stored; Step 2, analyzing the information quality of any image information in the image information of the animation to be stored, and obtaining the information evaluation value of the image information; comparing the information evaluation value with its quality threshold value, if it is higher than its quality threshold value, it means that the quality of the image information does not meet the conditions, and a re-execution mode is formed; otherwise, if it is not higher than the quality threshold value, it means that the quality of the image information meets the conditions; Step 3, evaluating the performance of the external memory storage area and performing improvement treatment, obtaining the performance evaluation value of the external memory storage area, and comparing it with the corresponding pre-defined critical quantity to obtain the corresponding performance treatment mode; Step 4, according to the image information of the animation, the storage unit required for the information is analyzed to obtain the information attribute quantity; the numerical range of the low, medium and high attributes of the image information is set, and the information attribute quantity is adapted to the numerical range of the low, medium and high attributes to obtain the attributes of the corresponding image information, and the corresponding storage unit is adapted according to the attributes of the image information; Step 5, evaluating the stability of each storage area in the external memory, and obtaining the overall normal parameters of a random page in the storage unit of the storage area; Step 6, performs additional quantity calculation on the overall normal parameters and performance evaluation quantities to obtain the overall level quantity of the page; performs additional quantity calculation on the information evaluation quantity and information attribute quantity of the image information to be stored in the corresponding storage unit, and distributes the image information after obtaining the overall level quantity of the information.

2. The real-time facial animation control method according to claim 1, characterized in that: In step 4, the attributes include low information attributes, medium information attributes, and high information attributes; If it is a low information attribute, the image information is prepared to be stored in a storage unit in the SRAM area of ​​the external memory; If it is a medium information attribute, the image information is prepared to be stored in the DDRAM area of ​​the external memory and then migrated to the storage unit of the DRAM area after a defined period of time; if it is a high information attribute, the image information is prepared to be stored in the storage unit of the DDRAM area of ​​the external memory.

3. The real-time facial animation control method according to claim 2, characterized in that: In step 6, the method for distributing image information is executed, including: distributing the image information with the largest overall information level to the storage page with the largest overall information level of its corresponding storage unit; removing the distributed image information and pages from the queue to be distributed, and then distributing the next image information in accordance with the method of distributing the image information with the largest overall information level to the storage page with the largest overall information level of the corresponding storage unit, until all image information in the animation image information is distributed or there is no appropriate storage page.

4. The real-time facial animation control method according to claim 3, characterized in that: The method for analyzing the information quality of any image information in the image information of the animation to be stored comprises: Obtain information receiving information corresponding to the image information of the animation, the information receiving information including the information forming capacity and the information forming temporary storage capacity of the image information; confirm information piece information of any image information in the image information of the animation, the information piece information including the information transmission capacity and the information main storage capacity of the image information; The information formation capacity is the capacity when the image information is formed. The information formation temporary storage capacity is the capacity occupied by the temporary storage when the graphic information is formed and transmitted to the computer terminal. The information transmission capacity is the capacity when the image information is taken out of the temporary storage and is to be transmitted to the main memory of the computer terminal. The information main memory capacity is the capacity occupied by the image information when it is transmitted to the main memory.

5. The real-time facial animation control method according to claim 4, characterized in that: The method for analyzing the information quality of any image information in the image information of the animation to be stored also includes: Processing is performed on any image information in the image information of the animation corresponding to the image information of the animation, and the hash value of the random image information is defined as IG1 by using the hash method; the information formation capacity and the information formation temporary storage capacity of the random image information are respectively subtracted from the information transmission capacity and the information main memory capacity, so as to obtain the information capacity spacing amount and the area spacing amount respectively, which are defined as IG2 and IG3 respectively; the hash value of the image information, the information capacity spacing amount and the area spacing amount are calculated, and the equation IG=IG1*i1+IG2*i2+IG3*i3 is used to obtain the information evaluation amount IG; here, i1, i2, i3 respectively represent the criticality coefficients corresponding to the pre-set hash value, the information capacity spacing amount and the area spacing amount.

6. The real-time facial animation control method according to claim 5, characterized in that: Evaluate the performance of external memory storage areas and implement methods to improve their handling, including: Confirm the performance parameters of the external memory storage area, perform spacing calculation on any parameter in the performance parameters and its corresponding base parameter, and obtain the performance spacing corresponding to the parameter, which is defined as H k , k represents the serial number of the parameter corresponding to the performance spacing quantity; The performance gap of all performance parameters in the external memory storage area is calculated using the equation Get performance evaluation measure IH1; here, A pre-set key quantity representing the performance gap quantity corresponding to the sequence number k; Set a corresponding pre-defined critical quantity for the performance evaluation, and compare the performance evaluation quantity with its corresponding pre-defined critical quantity to obtain a corresponding performance handling mode, that is, if the performance evaluation quantity is lower than its corresponding pre-defined critical quantity, a corresponding coping mode is formed.

7. The real-time facial animation control method according to claim 6, characterized in that: The method for parsing the storage unit of the information according to the image information of the animation includes: The browsing frequency of the same image information as any image information in the image information of the animation is defined as Lu, u represents the serial number of the same image information, and the operation is performed based on the browsing frequency of any image information and the average browsing frequency of the same image information, using the equation Obtain the browsing frequency measure LH of the same image information; here, Lu represents the browsing frequency of the image information corresponding to the sequence number u, represents the average browsing frequency of image information of the same category, and uδ represents the number The corresponding key quantity; Construct a task weight detection message table, send any image information of the animation image information into the task weight detection message table to obtain, send different detection methods to obtain the task weight of any image information and define it as Gs, s represents the serial number of the detection method contained in the task weight detection message table, and perform calculations on the task weight of any image information using different detection methods, using the equation GS = ∑ s＝1 (sGS*sι), obtain the weight level GS; here, sGS and sι represent the task weight corresponding to the detection method with sequence number s and the corresponding key quantity set in advance; Get the ease of acquisition information of the animated image information, and process the ease of acquisition information, that is, use the equation QH = ∑ q＝1 (qLH*qν), obtain the abstract value QH; The browsing frequency metric LH, the weight level GS, and the abstract value QH are quantified and the equation is used. The information attribute quantity TY is obtained; here, y1, y2, and y3 represent the pre-set key effect coefficients corresponding to the browsing frequency measurement, the weight level, and the abstract cost quantity, respectively.

8. The real-time facial animation control method according to claim 7, characterized in that: Methods for evaluating the stability of various storage areas in external memory include: Extract the residual storage capacity of each storage area in the external memory, and define the storage area whose residual storage capacity is higher than the storage capacity threshold value set in advance as the available storage area; set the maximum access and clearing frequency CCzg of the available storage area; obtain the access and clearing frequency CCxz of different pages corresponding to the storage units of each available storage area; use the equation CCt=CCzg-CCxz to subtract the access and clearing frequency from the maximum access and clearing frequency to obtain the residual access and clearing frequency CCt; The quotient obtained by dividing the frequency of virus attacks on each page by the frequency of page access and clearing is used as the page stability parameter HE; the quotient obtained by dividing the frequency of page failure by the frequency of page access and clearing is used to obtain the failure anomaly probability HT; the residual access and clearing frequency, stability coefficient, and failure anomaly probability are processed, and the equation HC = (CCt*e1+HE*e2) / (HT*e3+1) is used to obtain the overall normal parameter HC of the page; here, e1, e2, and e3 respectively represent the pre-set key quantities corresponding to the residual access and clearing frequency, stability coefficient, and failure anomaly probability.

9. The real-time facial animation control method according to claim 8, characterized in that: When forming an information re-execution model, it includes: Step 2-1, collect new image information again, then use the median filter method to filter the new image information and then perform step 2 to obtain a new information evaluation measure; Step 2-2, set the repetition frequency, and obtain a new information evaluation quantity for each repetition; Step 2-3, compare the maximum new information evaluation value with the quality threshold value set in advance. If it is higher than the quality threshold value, it means that the quality of the image information does not meet the requirements, and a re-selection process is formed; otherwise, if it is not higher than the quality threshold value, it means that the quality of the image information meets the requirements; Step 2-4, when forming the re-taking fluctuation process, the new information evaluation quantity and the initial information evaluation quantity are subjected to the interval quantity operation to obtain the re-taking evaluation interval quantity; the re-taking evaluation interval quantity is subjected to the standard deviation operation to obtain the evaluation fluctuation value; the numerical interval of the evaluation frequency cluster is set, and the evaluation fluctuation value is adapted to the numerical interval of its evaluation frequency cluster to obtain the corresponding evaluation frequency; the evaluation frequency is used to increase the repetition frequency; Step 2-5, execute step 2-3 again. If the maximum new information evaluation value is still higher than its quality threshold, remove the image information and form a discarded information entry with all discarded image information. If the new information evaluation value is not higher than its quality threshold, it means that the quality of the image information meets the conditions.

10. A real-time facial animation control system, characterized in that: include: A computer terminal with an external memory connected to the image acquisition device, the image acquisition device is used to transmit the acquired face image to the computer terminal, the computer terminal is used to generate a facial animation using the face image as a key frame, and then the face image and the facial animation are stored in the external memory, the face image or the facial animation is the image information of the animation; The modules running on the computer terminal include: A receiving module, which is used to receive the image information of the animation to be stored; The analysis module is used to analyze the information quality of any image information in the image information of the animation to be stored, and obtain the information evaluation value of the image information; compare the information evaluation value with its quality threshold value, if it is higher than its quality threshold value, it means that the quality of the image information does not meet the conditions, and the information re-execution mode is formed; otherwise, if it is not higher than the quality threshold value, it means that the quality of the image information meets the conditions; An evaluation module is used to evaluate the performance of the external memory storage area and perform improvement treatment, obtain the performance evaluation value of the external memory storage area, and compare it with the corresponding pre-defined critical quantity to obtain the corresponding performance treatment mode; A setting module is used to analyze the storage unit required for the information according to the image information of the animation to obtain the information attribute quantity; set the numerical range of the low, medium and high attributes of the image information, and adapt the information attribute quantity to the numerical range of the low, medium and high attributes to obtain the attributes of the corresponding image information, and adapt the corresponding storage unit according to the attributes of the image information; A stability module is used to evaluate the stability of each storage area in the external memory and obtain the overall normal parameters of any page in the storage unit of the storage area; The distribution module is used to perform additional quantity operations on the overall normal parameters and performance evaluation quantities to obtain the overall level quantity of the page; perform additional quantity operations on the information evaluation quantities and information attribute quantities of the image information to be stored in the corresponding storage unit, and distribute the image information after obtaining the overall level quantity of the information.

Images