Frame image prediction method and device, computer device and storage medium

Abstract

This application relates to the field of visual coding technology, and in particular to a frame image prediction method, apparatus, computer device, and storage medium. The method includes: enhancing the brightness of an initial reference frame image to obtain an input reference frame image; predicting the output frame based on the input reference frame image to obtain an initial output frame image; and restoring the brightness of the initial output frame image to obtain a target output frame image. This application reduces unnecessary prediction iterations, improves the prediction efficiency of frame images, and ensures the effectiveness of frame image generation. Furthermore, by enhancing the brightness of the initial reference frame image, and then predicting the output frame based on the input reference frame image, it ensures that the initial output frame image is not affected by insufficient brightness of the initial reference frame image during the prediction process, further improving the prediction accuracy of the initial output frame image.

Description

[0001] Cross-reference to related applications

[0002] This application claims priority to Chinese Patent Application No. 202510368276.3, filed on March 26, 2025, entitled “Frame Image Prediction Method, Apparatus, Computer Equipment and Storage Medium”, the entire contents of which are incorporated herein by reference. Technical Field

[0003] This application relates to the field of visual coding technology, and in particular to a frame image prediction method, apparatus, computer device, and storage medium. Background Technology

[0004] Existing VCMs adopt temporal frame interpolation methods, including interpolation, which uses forward and backward reference frames to predict intermediate frames, and extrapolation, which uses forward reference frames to predict subsequent frames.

[0005] However, existing frame interpolation methods use a binary search approach for different frame interpolation ratios. For sampling ratios corresponding to the same binary tree level, the entire binary tree needs to be predicted before selection, which increases the number of unnecessary predictions. Summary of the Invention

[0006] Therefore, it is necessary to provide a frame image prediction method, apparatus, computer device, and storage medium that can reduce the number of predictions to address the above-mentioned technical problems.

[0007] Firstly, this application provides a frame image prediction method. The method includes:

[0008] The initial reference frame image is brightened to obtain the input reference frame image;

[0009] Based on the input reference frame image, perform frame image prediction on the output frame to obtain the initial output frame image;

[0010] The initial output frame image is brightness restored to obtain the target output frame image.

[0011] In one embodiment, the step of performing frame image prediction on the output frame based on the input reference frame image to obtain an initial output frame image includes:

[0012] Select the output frame index from the candidate frame indices;

[0013] Determine the frame image index required for frame image prediction of the output frame;

[0014] Frame image prediction is performed based on the input reference frame image to obtain the frame image corresponding to the frame image index;

[0015] Based on the frame image required for frame image prediction of the output frame, perform frame image prediction on the output frame to obtain the initial output frame image corresponding to the output frame index.

[0016] In one embodiment, the step of performing frame image prediction on the output frame based on the input reference frame image to obtain an initial output frame image further includes:

[0017] Based on the forward and backward prediction function, the output frame is predicted according to the input reference frame image to obtain the initial output frame image.

[0018] In one embodiment, selecting the output frame index from the candidate frame index includes:

[0019] Obtain the number of frames to be predicted;

[0020] Select the output frame index corresponding to the number of frames to be predicted from the candidate frame index.

[0021] In one embodiment, selecting the output frame index corresponding to the number of frames to be predicted from the candidate frame index includes:

[0022] The target positions of the frame images are determined at equal intervals based on the number of frames to be predicted, and the output frame index corresponding to the frame to be predicted that is closest to the target position is selected from the candidate frame index.

[0023] In one embodiment, when the image format of the initial reference frame image is a luminance-chrominance YUV format, the step of luminance enhancement of the initial reference frame image to obtain the input reference frame image includes:

[0024] The initial reference frame image is brightened to obtain the enhanced initial reference frame image;

[0025] The enhanced initial reference frame image is converted to RGB format to obtain the input reference frame image.

[0026] In one embodiment, the step of enhancing the brightness of the initial reference frame image to obtain an enhanced initial reference frame image includes:

[0027] Obtain the average brightness parameter corresponding to the initial reference frame image;

[0028] If the brightness quantization index of the initial reference frame image is less than the preset brightness enhancement trigger threshold, the brightness enhancement status flag of the initial reference frame image is set according to the average brightness parameter, and the brightness parameter of the initial reference frame image is shifted to the left to obtain the enhanced initial reference frame image.

[0029] In one embodiment, where the initial reference frame image is in YUV format, the step of enhancing the brightness of the initial reference frame image to obtain the input reference frame image includes:

[0030] The initial reference frame image is converted to RGB format to obtain the reference frame image after the first format conversion.

[0031] The brightness of the reference frame image after the first format conversion is enhanced to obtain the input reference frame image.

[0032] In one embodiment, the step of enhancing the brightness of the format-converted reference frame image to obtain the input reference frame image includes:

[0033] Perform YUV conversion on the reference frame image after the first format conversion to obtain the initial reference frame image after the second format conversion;

[0034] The brightness of the initial reference frame image after the second format conversion is enhanced to obtain the enhanced initial reference frame image.

[0035] The enhanced initial reference frame image is converted to RGB format to obtain the input reference frame image.

[0036] Secondly, this application also provides a frame image prediction apparatus. The apparatus includes:

[0037] The enhancement module is used to enhance the brightness of the initial reference frame image to obtain the input reference frame image;

[0038] The prediction module is used to predict the output frame based on the input reference frame image to obtain an initial output frame image;

[0039] The restoration module is used to restore the brightness of the initial output frame image to obtain the target output frame image.

[0040] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0041] The initial reference frame image is brightened to obtain the input reference frame image;

[0042] Based on the input reference frame image, perform frame image prediction on the output frame to obtain the initial output frame image;

[0043] The initial output frame image is brightness restored to obtain the target output frame image.

[0044] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0045] The initial reference frame image is brightened to obtain the input reference frame image;

[0046] Based on the input reference frame image, perform frame image prediction on the output frame to obtain the initial output frame image;

[0047] The initial output frame image is brightness restored to obtain the target output frame image.

[0048] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, performs the following steps:

[0049] The initial reference frame image is brightened to obtain the input reference frame image;

[0050] Based on the input reference frame image, perform frame image prediction on the output frame to obtain the initial output frame image;

[0051] The initial output frame image is brightness restored to obtain the target output frame image.

[0052] The aforementioned frame image prediction method, apparatus, computer device, and storage medium obtain an input reference frame image by enhancing the brightness of an initial reference frame image, and then predict the output frame based on the input reference frame image to obtain an initial output frame image; furthermore, the initial output frame image is brightness restored to obtain a target output frame image. As can be seen from the above, the frame image prediction process of this application does not require constructing a complete binary tree structure; instead, it only needs to perform targeted prediction of the required frame image based on the input reference frame image to obtain the initial output frame image. This reduces unnecessary prediction iterations, improves the prediction efficiency of the frame image, and ensures the effectiveness of frame image generation. Furthermore, by enhancing the brightness of the initial reference frame image and then predicting the output frame based on the input reference frame image, the generation quality of the initial output frame image is not affected by the low brightness of the initial reference frame image, further improving the prediction accuracy of the initial output frame image. Attached Figure Description

[0053] Figure 1 An application environment diagram of a frame image prediction method provided in this application embodiment;

[0054] Figure 2 A flowchart illustrating the first frame image prediction method provided in this application embodiment;

[0055] Figure 3 A flowchart illustrating the second frame image prediction method provided in this application embodiment;

[0056] Figure 4 A flowchart illustrating the third frame image prediction method provided in this application embodiment;

[0057] Figure 5 A flowchart illustrating the fourth frame image prediction method provided in this application embodiment;

[0058] Figure 6 A flowchart illustrating the fifth frame image prediction method provided in this application embodiment;

[0059] Figure 7 A flowchart illustrating the sixth frame image prediction method provided in this application embodiment;

[0060] Figure 8 A flowchart illustrating the seventh frame image prediction method provided in this application embodiment;

[0061] Figure 9 A structural block diagram of a frame image prediction device provided in an embodiment of this application;

[0062] Figure 10 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0063] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0064] The frame image prediction method provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or placed on a cloud or other network server. An input reference frame image is obtained by enhancing the brightness of an initial reference frame image; then, frame image prediction is performed on the output frame based on the input reference frame image to obtain an initial output frame image; subsequently, brightness restoration is performed on the initial output frame image to obtain the target output frame image. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. Portable wearable devices can include smartwatches, smart bracelets, head-mounted devices, etc. Server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0065] In one embodiment, the temporal frame interpolation module called by the frame image prediction method uses a binary search method to call the image prediction module to generate candidate frame images.

[0066] In one embodiment, such as Figure 2 As shown, a frame image prediction method is provided, which is applied to... Figure 1 Taking server 104 as an example, the following steps are included:

[0067] S201, Brightness enhancement is performed on the initial reference frame image to obtain the input reference frame image.

[0068] It should be noted that when it is necessary to enhance the brightness of the initial reference frame image, the image format of the initial reference frame image can be determined in advance. If the image format of the initial reference frame image is YUV format, the brightness of the initial reference frame image can be directly enhanced to obtain the input reference frame image. If the image format of the initial reference frame image is not YUV format, the format of the initial reference frame image is changed, and the brightness of the changed YUV format initial reference frame image is enhanced to obtain the input reference frame image.

[0069] To further explain, when enhancing the brightness of the initial reference frame image, the average brightness parameter corresponding to the initial reference frame image can be obtained. Then, the brightness enhancement status flag of the initial reference frame image is set, and the brightness parameter of the initial reference frame image is shifted to the left. If the average brightness parameter is greater than the preset brightness, the brightness enhancement is determined to be complete, and the enhanced initial reference frame image is obtained.

[0070] S202, perform frame image prediction on the output frame based on the input reference frame image to obtain the initial output frame image.

[0071] It should be noted that when it is necessary to perform frame image prediction on the output frame based on the input reference frame image, the following steps may be included: selecting the output frame index from the candidate frame index; determining the frame image index required for frame image prediction on the output frame; performing frame image prediction based on the input reference frame image to obtain the frame image corresponding to the frame image index; and then performing frame image prediction on the output frame based on the frame image required for frame image prediction on the output frame to obtain the initial output frame image corresponding to the output frame index.

[0072] S203, perform brightness restoration on the initial output frame image to obtain the target output frame image.

[0073] It should be noted that when it is necessary to restore the brightness of the initial output frame image, the format of the initial output frame image is changed to ensure that the image format of the initial output frame image is changed to the YUV format that can be used for brightness restoration. Then, after the format change is completed, the brightness of the initial output frame image is restored to obtain the target output frame image.

[0074] Specifically, when performing brightness restoration on the initial output frame image, the following can be included: based on the forward and backward prediction functions, perform frame image prediction on the output frame according to the input reference frame image to obtain the initial output frame image.

[0075] The aforementioned frame image prediction method obtains an input reference frame image by enhancing the brightness of an initial reference frame image, and then predicts the output frame based on the input reference frame image to obtain an initial output frame image. Finally, it restores the brightness of the initial output frame image to obtain the target output frame image. As can be seen from the above, this application does not require constructing a complete binary tree structure during frame image prediction. Instead, it only needs to perform targeted prediction of the required frame image based on the input reference frame image to obtain the initial output frame image. This reduces unnecessary prediction iterations, improves the prediction efficiency of frame images, and ensures the effectiveness of frame image generation. Furthermore, by enhancing the brightness of the initial reference frame image and then predicting the output frame based on the input reference frame image, it ensures that the initial output frame image is not affected by low brightness of the initial reference frame image during prediction, further improving the prediction accuracy of the initial output frame image.

[0076] In one embodiment, such as Figure 3 As shown, when it is necessary to predict the output frame based on the input reference frame image to obtain the initial output frame image, the following can be included:

[0077] S301, Select the output frame index from the candidate frame index.

[0078] It should be noted that when it is necessary to select the output frame index from the candidate frame index, the following can be included: obtaining the number of frames to be predicted; selecting the output frame index corresponding to the number of frames to be predicted from the candidate frame index.

[0079] The number of frames to be predicted is used to characterize the number of frames dropped between two known frames. Specifically, the number of frames to be predicted can be determined according to the following calculation formula.

[0080] ;

[0081] Where d refers to the number of frames to be predicted; S refers to the number of binary prediction iterations; and r refers to the calculation coefficient corresponding to 2.

[0082] To further explain, when it is necessary to select the output frame index corresponding to the number of frames to be predicted from the candidate frame index, the following can be included: determine the target positions of the frame images at equal intervals according to the number of frames to be predicted, and select the output frame index corresponding to the frame to be predicted that is closest to the target position from the candidate frame index.

[0083] As an example, if it is determined that there are 15 candidate frame indices, two reference frame indices are 0 and 16, and the candidate frame indices are 1, 2, ..., 15; where the total length between the two reference frames is taken as 1, the position corresponding to each candidate frame index can be represented as 1 / 16, 2 / 16, 3 / 16...15 / 16; if the number of frames to be predicted is 9, the target positions of the frames to be predicted at equal intervals are 1 / 10, 2 / 10...9 / 10; after determining the target position, if the target position is represented as 1 / 10 under the total length of the 15 candidate frame indices, then the candidate frame index 2, which is closest to 1 / 10, is taken as the output frame index.

[0084] S302, determine the frame image index required for frame image prediction of the output frame.

[0085] It should be noted that when it is necessary to determine the frame image index required for frame image prediction of the output frame, the operation of determining the frame image index required for frame image prediction of the output frame can be realized based on the relationship between the output frame index and the frame index.

[0086] Among them, the frame image required for the predicted frame image index is the input reference frame image.

[0087] The frame index association relationship records at least one frame index, which is used to predict the frame image index of the frame image required for the corresponding frame image. As an example, the frame index association relationship may record:

[0088] {8:[0,16],4:[0,8],12:[8,16],2:[0,4],6:[4,8],10:[8,12],14:[12,16],1:[0,2],3:[2,4],5:[4,6],7:[6,8],9:[8,10],11:[10,12],13:[12,14],15:[14,16]};

[0089] Taking 8:[0,16] as an example, the frame image indices required for frame image prediction of the output frame index 8 are frame image index 0 and frame image index 16.

[0090] S303, perform frame image prediction based on the input reference frame image to obtain the frame image corresponding to the frame image index.

[0091] It should be noted that since the frame image required to predict the frame image corresponding to the frame image index is the input reference frame image, the frame image can be predicted based on the input reference frame image to obtain the frame image corresponding to the frame image index.

[0092] In one embodiment of this application, if the output frame index is 4 and the reference frame indices of the input reference frame image are 0 and 16, then according to the frame index association, the frame image index required for frame image prediction of the output frame is determined to be 8. Therefore, based on the input reference frame images with frame image indices of 0 and 16, frame image prediction is performed to obtain the frame image corresponding to frame image index 8. Subsequently, the initial output frame image can be generated based on the frame image corresponding to frame image index 8 and the input reference frame image with frame image index 0.

[0093] S304, perform frame image prediction on the output frame based on the frame image required for frame image prediction on the output frame, and obtain the initial output frame image corresponding to the output frame index.

[0094] It should be noted that when frame image prediction is required, the initial output frame image can be obtained by performing frame image prediction based on the output frame index, using the forward and backward prediction functions and the frame image required for frame image prediction of the output frame.

[0095] In one embodiment of this application, if there are a total of 16 candidate frame indices, when a frame image with frame index 1 needs to be generated, it is determined according to the frame index association relationship that generating a frame image with frame index 1 requires frame images with frame indices 0 and 2; however, if a frame image with frame index 2 is missing, it is determined according to the frame index association relationship that generating a frame image with frame index 2 requires frame images with frame indices 0 and 4; however, if a frame image with frame index 4 is missing, it is determined according to the frame index association relationship that generating a frame image with frame index 4 requires frame images with frame indices 0 and 8; based on the frame index association relationship, a frame image with frame index 8 is generated from the frame images with frame indices 0 and 16, a frame image with frame index 4 is generated from the frame images with frame indices 0 and 8, and a frame image with frame index 2 is generated from the frame images with frame indices 0 and 4; finally, a frame image with index 1 is generated from the frame images with frame indices 0 and 2.

[0096] The above-described frame image prediction method, by determining the output frame index and the frame image corresponding to the frame image index, enables the prediction of the output frame based on the frame image required for the frame image prediction of the output frame, thereby obtaining the initial output frame image corresponding to the output frame index. This method achieves the goal of only needing to perform targeted prediction of the required frame image based on the input reference frame image to obtain the initial output frame image. It reduces unnecessary prediction times, improves the prediction efficiency of the frame image, and ensures the effectiveness of the generated frame image.

[0097] In one embodiment, such as Figure 4 As shown, when the initial reference frame image is in luminance / chrominance YUV format, and it is necessary to enhance the luminance of the initial reference frame image to obtain the input reference frame image, the following can be included:

[0098] S401, the brightness of the initial reference frame image is enhanced to obtain the enhanced initial reference frame image.

[0099] It should be noted that when it is necessary to enhance the brightness of the initial reference frame image to obtain the enhanced initial reference frame image, the following can be included: obtaining the average brightness parameter corresponding to the initial reference frame image; if the brightness quantization index of the initial reference frame image is less than the preset brightness enhancement trigger threshold, according to the average brightness parameter, setting the brightness enhancement status flag of the initial reference frame image, shifting the brightness parameter of the initial reference frame image to the left, and obtaining the enhanced initial reference frame image.

[0100] If the average brightness parameter is determined to be much greater than the preset brightness (e.g., average brightness parameter >> 2), then the initial reference frame image enhancement is determined to be successful.

[0101] S402, perform RGB format conversion on the enhanced initial reference frame image to obtain the input reference frame image.

[0102] In one embodiment of this application, such as Figure 5 As shown, when it is necessary to predict the target output frame image, the following steps may be included: obtaining the average brightness parameter corresponding to the initial reference frame image; if the brightness quantization index of the initial reference frame image is less than a preset brightness enhancement trigger threshold, adjusting the brightness enhancement status flag of the initial reference frame image according to the average brightness parameter, and shifting the brightness parameter of the initial reference frame image to the left to obtain the enhanced initial reference frame image; converting the enhanced initial reference frame image to RGB format to obtain the input reference frame image; performing frame image prediction on the output frame based on the input reference frame image to obtain the initial output frame image; converting the initial output frame image from RGB format to YUV format, restoring the brightness enhancement status flag of the initial output frame image in YUV format, and shifting the brightness parameter of the initial reference frame image to the right to achieve brightness restoration, thus obtaining the target output frame image.

[0103] The above-mentioned frame image prediction method enhances the brightness of the initial reference frame image to obtain an enhanced initial reference frame image; and converts the enhanced initial reference frame image to RGB format to obtain an input reference frame image; this ensures that the initial output frame image will not be affected by the low brightness of the initial reference frame image during the prediction generation process, thereby further improving the prediction accuracy of the initial output frame image.

[0104] In one embodiment, such as Figure 6 As shown, when the initial reference frame image is in luminance / chrominance YUV format, and it is necessary to enhance the luminance of the initial reference frame image to obtain the input reference frame image, the following can be included:

[0105] S601, the initial reference frame image is converted to RGB format to obtain the reference frame image after the first format conversion.

[0106] S602, the brightness of the reference frame image after the first format conversion is enhanced to obtain the input reference frame image.

[0107] It should be noted that when it is necessary to enhance the brightness of the format-converted reference frame image to obtain the input reference frame image, the following may be included: performing YUV conversion on the format-converted reference frame image to obtain the initial reference frame image after the second format conversion; enhancing the brightness of the initial reference frame image after the second format conversion to obtain the enhanced initial reference frame image; and performing RGB format conversion on the enhanced initial reference frame image to obtain the input reference frame image.

[0108] In one embodiment of this application, such as Figure 7 As shown, the initial reference frame image is converted to RGB format to obtain a reference frame image after the first format conversion. The reference frame image after the first format conversion is then converted to YUV format to obtain an initial reference frame image after the second format conversion. The brightness enhancement status flag of the initial reference frame image after the second format conversion is set, and the brightness parameter of the enhanced initial reference frame image is shifted to the left to obtain the initial reference frame image after the second format conversion (before and after enhancement). The initial reference frame image after the second format conversion (before and after enhancement) is then converted from YUV to RGB format to obtain the input reference frame image. Frame image prediction is performed on the output frame based on the input reference frame image to obtain the initial output frame image. The initial output frame image is then converted from RGB format to YUV format, and the brightness enhancement status flag of the initial output frame image in YUV format is restored. The brightness parameter of the initial reference frame image is shifted to the right to achieve brightness restoration, resulting in the initial output frame image after brightness restoration. The initial output frame image after brightness restoration is then converted from YUV format to RGB format. Furthermore, if a target output frame image in YUV format is required, the initial output frame image in RGB format needs to be converted to YUV format to obtain the target output frame image.

[0109] The aforementioned frame image prediction method converts the initial reference frame image to RGB format to obtain a first-format converted reference frame image; then, it enhances the brightness of the first-format converted reference frame image to obtain the input reference frame image. This ensures that the initial output frame image is not affected by the low brightness of the initial reference frame image during the prediction generation process, further improving the prediction accuracy of the initial output frame image.

[0110] In one embodiment, the process of determining the target location based on the sampling rate includes: determining the target location based on the sampling rate;

[0111] Specifically, the sampling rate can be determined using an adaptive method.

[0112] In one embodiment of this application, when it is necessary to determine the sampling rate, the following may be included: setting the step size to 1, initializing i to 1, and initializing j to 4. Statistically calculating the frame difference between the i-th and j-th foreground targets, the area ratio of the foreground targets, and the average brightness of the foreground targets. The statistical method is as follows: First, using a target detection algorithm to identify the target boxes of the two frames of images, and using the target boxes as a mask layer to remove the background, setting the background pixels to 0, obtaining an image containing only the foreground targets, and statistically calculating the average brightness of the foreground targets and the area ratio of the foreground targets. Second, calculating the frame difference between the two foreground target images and converting them to grayscale images, statistically calculating the number of pixels in the frame difference image whose pixel values ​​are greater than a threshold (average brightness of the foreground targets * 30%), and dividing the number of pixels greater than the threshold by the total number of pixels in the foreground targets to obtain the frame difference of the foreground targets. The frame difference threshold is set to 0.4. When the average brightness of the foreground targets is lower than a certain threshold (80), the image is too dark, and the frame difference threshold needs to be lowered to 0.3. When the frame difference between two frames is greater than the frame difference threshold, or the area ratio of the foreground target is greater than 0.2, or ji >= the highest sampling rate, it is considered that the two frames have a significant difference or the foreground target is complex. In this case, (ji) is selected as the sampling rate for the i-th frame. The next step is to count the j-th and j+3-th frames. If the statistical values ​​of the two frames do not meet the conditions described in step 4, the next step is to count the i-th and (j+step 1)-th frames. This process continues until the entire video sequence is traversed, resulting in a set of sampling rates.

[0113] In another embodiment of this application, when it is necessary to determine the sampling rate, the following may be included: setting the step size to 1, initializing i to 1, and initializing j to 5. Statistically calculating the frame difference between the i-th and j-th foreground targets, the average brightness of the foreground targets, and the number of foreground targets. Statistical method: First, using a target detection algorithm to identify the target boxes of the two frames of images, and using the target boxes as a mask layer to remove the background, setting the background pixels to 0, obtaining an image containing only foreground targets, and statistically calculating the average brightness of the foreground targets, the area ratio of the foreground targets, and the number of foreground targets. Second, calculating the frame difference between the two foreground target images and converting them to grayscale images, statistically calculating the number of pixels in the frame difference image whose pixel value is greater than a threshold (average brightness of the foreground targets * 30%), and dividing the number of pixels greater than the threshold by the total number of pixels in the foreground targets to obtain the frame difference of the foreground targets. The frame difference threshold is set to 0.39. When the average brightness of the foreground targets is lower than a certain threshold (80), the image is too dark, and the frame difference threshold needs to be lowered to 0.2. When the frame difference obtained from two frames is greater than the frame difference threshold, or the number of foreground targets is greater than the threshold (35), or ji>= the highest sampling rate, it is considered that the two frames have a large difference or the foreground targets are complex. Next, the frame difference is further judged. When the frame difference is greater than 0.59, it is considered that the difference between the two frames is too large, and the sampling rate needs to be adjusted back. The step size of the adjustment is 1, that is, the frame difference between the i-th frame and the (j-1)-th frame is counted until the minimum sampling rate is reached or the frame difference is no longer greater than 0.59. In addition, when the frame difference is not greater than 0.59, (ji) is selected as the sampling rate of the i-th frame. The next step is to count the j-th frame and the (j+4)-th frame. When the statistical values ​​of the two frames do not meet the conditions described in step 4, the next step is to count the i-th frame and the (j+step 1)-th frame. Until the entire video sequence is traversed, a set of sampling rates is obtained.

[0114] In one embodiment, such as Figure 8 As shown, when it is necessary to obtain the target output frame image, the following can be included:

[0115] S801, the brightness of the initial reference frame image is enhanced to obtain the input reference frame image.

[0116] S802, obtain the number of frames to be predicted.

[0117] S803: Determine the target positions at equal intervals in the frame images based on the number of frames to be predicted, and select the output frame index corresponding to the frame to be predicted that is closest to the target position from the candidate frame index.

[0118] S804, determine the frame image index required for frame image prediction of the output frame.

[0119] S805: Perform frame image prediction based on the input reference frame image to obtain the frame image corresponding to the frame image index.

[0120] S806, perform frame image prediction on the output frame based on the frame image required for frame image prediction on the output frame, and obtain the initial output frame image corresponding to the output frame index.

[0121] S807 performs brightness restoration on the initial output frame image to obtain the target output frame image.

[0122] In one embodiment, when it is necessary to perform frame image prediction on the output frame based on the input reference frame image to obtain an initial output frame image, the method may include: calling a time-domain interpolation module to perform frame image prediction on the output frame based on the input reference frame image to obtain an initial output frame image.

[0123] Furthermore, when it is necessary to call the temporal interpolation module to predict the output frame based on the input reference frame image to obtain the initial output frame image, it may include:

[0124] The temporal interpolation module includes an image interpolation module to generate several candidate frame images; the output image of the image interpolation module is a single candidate frame image, and the input image is a reference frame image of the output candidate frame image; the reference frame image of the output candidate frame image is the input reference frame image and / or the candidate frame image; the initial output frame image is selected from the candidate frame images.

[0125] The candidate frame images output by the image interpolation module are stored in the empty image storage buffer of the decoded image buffer and marked as being used for temporal reference.

[0126] The reference frame image for the output candidate frame image is the input reference frame image and / or the candidate frame image.

[0127] In one embodiment of this application, the input image of the image interpolation module is used as the reference frame image of the output candidate frame image;

[0128] Candidate frame images that are not part of the reference frame images to be generated are marked as not to be used for temporal reference;

[0129] Candidate frame images marked as not used for temporal reference and not as initial output frame images are removed from the decoded image buffer.

[0130] The reference frame image for the output candidate frame image is the input reference frame image and / or the candidate frame image;

[0131] In another embodiment of this application, the time when the candidate frame image, which is the initial output frame image, is output from the decoded image buffer is determined by the time when the input reference frame image is output from the decoded image buffer, the temporal sampling rate, and the order of the candidate frame images among the input reference frame images.

[0132] As an example, when the TrPictureOutputFlag of candidate frame image i is 1, its DPB output time dppoutputtime [n + i] is derived as follows:

[0133] DpbOutputTime[ n + i ] =DpbOutputTime[ n ] +

[0134] i * (DpbOutputTime[ n + TrRatio ] − DpbOutputTime[ n ]) / TrRatio

[0135] Here, image n is the first input image of the temporal interpolation module. TrRatio is the temporal sampling ratio between the first and second input images. For example, a TrRatio of 8 indicates that the number of output images between the first and second input images is 7. Candidate frame image i and the current candidate frame image are the i-th candidate frame image after the first input image between the first and second input images of the temporal interpolation module, where i is a positive integer less than TrAltPicCnt.

[0136] TrAltPicCnt = 2**ceil(log(2, TrRatio)).

[0137] The aforementioned frame image prediction method obtains an input reference frame image by enhancing the brightness of an initial reference frame image, and then predicts the output frame based on the input reference frame image to obtain an initial output frame image. Finally, it restores the brightness of the initial output frame image to obtain the target output frame image. As can be seen from the above, this application does not require constructing a complete binary tree structure during frame image prediction. Instead, it only needs to perform targeted prediction of the required frame image based on the input reference frame image to obtain the initial output frame image. This reduces unnecessary prediction iterations, improves the prediction efficiency of frame images, and ensures the effectiveness of frame image generation. Furthermore, by enhancing the brightness of the initial reference frame image and then predicting the output frame based on the input reference frame image, it ensures that the initial output frame image is not affected by low brightness of the initial reference frame image during prediction, further improving the prediction accuracy of the initial output frame image.

[0138] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0139] Based on the same inventive concept, this application also provides a frame image prediction apparatus for implementing the frame image prediction method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more frame image prediction apparatus embodiments provided below can be found in the limitations of the frame image prediction method described above, and will not be repeated here.

[0140] In one embodiment, such as Figure 9 As shown, a frame image prediction device is provided, comprising: an enhancement module 10, a prediction module 20, and a restoration module 30, wherein:

[0141] The enhancement module 10 is used to enhance the brightness of the initial reference frame image to obtain the input reference frame image.

[0142] The prediction module 20 is used to predict the output frame based on the input reference frame image to obtain the initial output frame image.

[0143] The restoration module 30 is used to restore the brightness of the initial output frame image to obtain the target output frame image.

[0144] In one embodiment, the output frame index is selected from the candidate frame index;

[0145] Determine the frame image index required for frame image prediction of the output frame;

[0146] Frame image prediction is performed based on the input reference frame image to obtain the frame image corresponding to the frame image index;

[0147] Based on the frame image required for frame image prediction of the output frame, perform frame image prediction on the output frame to obtain the initial output frame image corresponding to the output frame index.

[0148] In one embodiment, based on a forward and backward prediction function, the output frame is predicted according to the input reference frame image to obtain an initial output frame image.

[0149] In one embodiment, the target location is determined based on the sampling rate.

[0150] In one embodiment, the number of frames to be predicted is obtained;

[0151] Select the output frame index corresponding to the number of frames to be predicted from the candidate frame index.

[0152] In one embodiment, target positions at equal intervals in the frame images are determined based on the number of frames to be predicted, and the output frame index corresponding to the frame to be predicted closest to the target position is selected from the candidate frame index.

[0153] In one embodiment, the initial reference frame image is brightness enhanced to obtain an enhanced initial reference frame image;

[0154] The enhanced initial reference frame image is converted to RGB format to obtain the input reference frame image.

[0155] In one embodiment, the average brightness parameter corresponding to the initial reference frame image is obtained;

[0156] If the brightness quantization index of the initial reference frame image is less than the preset brightness enhancement trigger threshold, the brightness enhancement status flag of the initial reference frame image is set according to the average brightness parameter, and the brightness parameter of the initial reference frame image is shifted to the left to obtain the enhanced initial reference frame image.

[0157] In one embodiment, the initial reference frame image is converted to RGB format to obtain a reference frame image after the first format conversion;

[0158] The brightness of the reference frame image after the first format conversion is enhanced to obtain the input reference frame image.

[0159] In one embodiment, the reference frame image after the first format conversion is converted to YUV to obtain the initial reference frame image after the second format conversion.

[0160] The brightness of the initial reference frame image after the second format conversion is enhanced to obtain the enhanced initial reference frame image.

[0161] The enhanced initial reference frame image is converted to RGB format to obtain the input reference frame image.

[0162] The aforementioned frame image prediction device obtains an input reference frame image by enhancing the brightness of an initial reference frame image, and then predicts the output frame based on the input reference frame image to obtain an initial output frame image. Finally, it restores the brightness of the initial output frame image to obtain the target output frame image. As can be seen from the above, this application does not require constructing a complete binary tree structure during frame image prediction. Instead, it only needs to perform targeted prediction of the required frame image based on the input reference frame image to obtain the initial output frame image. This reduces unnecessary prediction attempts, improves the prediction efficiency of frame images, and ensures the effectiveness of frame image generation. Furthermore, by enhancing the brightness of the initial reference frame image and then predicting the output frame based on the input reference frame image, it ensures that the initial output frame image is not affected by low brightness of the initial reference frame image during prediction, further improving the prediction accuracy of the initial output frame image.

[0163] Each module in the aforementioned frame image prediction device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0164] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 10 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a frame image prediction method. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0165] Those skilled in the art will understand that Figure 10 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0166] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0167] The initial reference frame image is brightened to obtain the input reference frame image;

[0168] Based on the input reference frame image, perform frame image prediction on the output frame to obtain the initial output frame image;

[0169] The initial output frame image is brightness restored to obtain the target output frame image.

[0170] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0171] Select the output frame index from the candidate frame indices;

[0172] Determine the frame image index required for frame image prediction of the output frame;

[0173] Frame image prediction is performed based on the input reference frame image to obtain the frame image corresponding to the frame image index;

[0174] Based on the frame image required for frame image prediction of the output frame, perform frame image prediction on the output frame to obtain the initial output frame image corresponding to the output frame index.

[0175] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0176] Based on the forward and backward prediction function, the output frame is predicted according to the input reference frame image to obtain the initial output frame image.

[0177] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0178] Obtain the number of frames to be predicted;

[0179] Select the output frame index corresponding to the number of frames to be predicted from the candidate frame index.

[0180] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0181] The target positions of the frame images are determined at equal intervals based on the number of frames to be predicted, and the output frame index corresponding to the frame to be predicted that is closest to the target position is selected from the candidate frame index.

[0182] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0183] The initial reference frame image is brightened to obtain the enhanced initial reference frame image;

[0184] The enhanced initial reference frame image is converted to RGB format to obtain the input reference frame image.

[0185] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0186] Obtain the average brightness parameter corresponding to the initial reference frame image;

[0187] If the brightness quantization index of the initial reference frame image is less than the preset brightness enhancement trigger threshold, the brightness enhancement status flag of the initial reference frame image is set according to the average brightness parameter, and the brightness parameter of the initial reference frame image is shifted to the left to obtain the enhanced initial reference frame image.

[0188] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0189] The initial reference frame image is converted to RGB format to obtain the reference frame image after the first format conversion.

[0190] The brightness of the reference frame image after the first format conversion is enhanced to obtain the input reference frame image.

[0191] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0192] Perform YUV conversion on the reference frame image after the first format conversion to obtain the initial reference frame image after the second format conversion;

[0193] The brightness of the initial reference frame image after the second format conversion is enhanced to obtain the enhanced initial reference frame image.

[0194] The enhanced initial reference frame image is converted to RGB format to obtain the input reference frame image.

[0195] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0196] The initial reference frame image is brightened to obtain the input reference frame image;

[0197] Based on the input reference frame image, perform frame image prediction on the output frame to obtain the initial output frame image;

[0198] The initial output frame image is brightness restored to obtain the target output frame image.

[0199] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0200] Select the output frame index from the candidate frame indices;

[0201] Determine the frame image index required for frame image prediction of the output frame;

[0202] Frame image prediction is performed based on the input reference frame image to obtain the frame image corresponding to the frame image index;

[0203] Based on the frame image required for frame image prediction of the output frame, perform frame image prediction on the output frame to obtain the initial output frame image corresponding to the output frame index.

[0204] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0205] Based on the forward and backward prediction function, the output frame is predicted according to the input reference frame image to obtain the initial output frame image.

[0206] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0207] Obtain the number of frames to be predicted;

[0208] Select the output frame index corresponding to the number of frames to be predicted from the candidate frame index.

[0209] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0210] The target positions of the frame images are determined at equal intervals based on the number of frames to be predicted, and the output frame index corresponding to the frame to be predicted that is closest to the target position is selected from the candidate frame index.

[0211] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0212] The initial reference frame image is brightened to obtain the enhanced initial reference frame image;

[0213] The enhanced initial reference frame image is converted to RGB format to obtain the input reference frame image.

[0214] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0215] Obtain the average brightness parameter corresponding to the initial reference frame image;

[0216] If the brightness quantization index of the initial reference frame image is less than the preset brightness enhancement trigger threshold, the brightness enhancement status flag of the initial reference frame image is set according to the average brightness parameter, and the brightness parameter of the initial reference frame image is shifted to the left to obtain the enhanced initial reference frame image.

[0217] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0218] The initial reference frame image is converted to RGB format to obtain the reference frame image after the first format conversion.

[0219] The brightness of the reference frame image after the first format conversion is enhanced to obtain the input reference frame image.

[0220] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0221] Perform YUV conversion on the reference frame image after the first format conversion to obtain the initial reference frame image after the second format conversion;

[0222] The brightness of the initial reference frame image after the second format conversion is enhanced to obtain the enhanced initial reference frame image.

[0223] The enhanced initial reference frame image is converted to RGB format to obtain the input reference frame image.

[0224] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0225] The initial reference frame image is brightened to obtain the input reference frame image;

[0226] Based on the input reference frame image, perform frame image prediction on the output frame to obtain the initial output frame image;

[0227] The initial output frame image is brightness restored to obtain the target output frame image.

[0228] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0229] Select the output frame index from the candidate frame indices;

[0230] Determine the frame image index required for frame image prediction of the output frame;

[0231] Frame image prediction is performed based on the input reference frame image to obtain the frame image corresponding to the frame image index;

[0232] Based on the frame image required for frame image prediction of the output frame, perform frame image prediction on the output frame to obtain the initial output frame image corresponding to the output frame index.

[0233] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0234] Based on the forward and backward prediction function, the output frame is predicted according to the input reference frame image to obtain the initial output frame image.

[0235] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0236] Obtain the number of frames to be predicted;

[0237] Select the output frame index corresponding to the number of frames to be predicted from the candidate frame index.

[0238] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0239] The target positions of the frame images are determined at equal intervals based on the number of frames to be predicted, and the output frame index corresponding to the frame to be predicted that is closest to the target position is selected from the candidate frame index.

[0240] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0241] The initial reference frame image is brightened to obtain the enhanced initial reference frame image;

[0242] The enhanced initial reference frame image is converted to RGB format to obtain the input reference frame image.

[0243] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0244] Obtain the average brightness parameter corresponding to the initial reference frame image;

[0245] If the brightness quantization index of the initial reference frame image is less than the preset brightness enhancement trigger threshold, the brightness enhancement status flag of the initial reference frame image is set according to the average brightness parameter, and the brightness parameter of the initial reference frame image is shifted to the left to obtain the enhanced initial reference frame image.

[0246] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:

[0247] The initial reference frame image is converted to RGB format to obtain the reference frame image after the first format conversion.

[0248] The brightness of the reference frame image after the first format conversion is enhanced to obtain the input reference frame image.

[0249] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0250] Perform YUV conversion on the reference frame image after the first format conversion to obtain the initial reference frame image after the second format conversion;

[0251] The brightness of the initial reference frame image after the second format conversion is enhanced to obtain the enhanced initial reference frame image.

[0252] The enhanced initial reference frame image is converted to RGB format to obtain the input reference frame image.

[0253] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0254] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0255] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0256] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A frame image prediction method, characterized in that, The method includes: The initial reference frame image is brightened to obtain the input reference frame image; Based on the input reference frame image, perform frame image prediction on the output frame to obtain the initial output frame image; The initial output frame image is brightness restored to obtain the target output frame image.

2. The method according to claim 1, characterized in that, The step of performing frame image prediction on the output frame based on the input reference frame image to obtain the initial output frame image includes: Select the output frame index from the candidate frame indices; Determine the frame image index required for frame image prediction of the output frame; Frame image prediction is performed based on the input reference frame image to obtain the frame image corresponding to the frame image index; Based on the frame image required for frame image prediction of the output frame, perform frame image prediction on the output frame to obtain the initial output frame image corresponding to the output frame index.

3. The method according to claim 1, characterized in that, The step of performing frame image prediction on the output frame based on the input reference frame image to obtain the initial output frame image further includes: Based on the forward and backward prediction function, the output frame is predicted according to the input reference frame image to obtain the initial output frame image.

4. The method according to claim 2, characterized in that, The step of selecting the output frame index from the candidate frame index includes: Obtain the number of frames to be predicted; Select the output frame index corresponding to the number of frames to be predicted from the candidate frame index.

5. The method according to claim 4, characterized in that, The step of selecting the output frame index corresponding to the number of frames to be predicted from the candidate frame index includes: The target positions of the frame images are determined at equal intervals based on the number of frames to be predicted, and the output frame index corresponding to the frame to be predicted that is closest to the target position is selected from the candidate frame index.

6. The method according to claim 1, characterized in that, When the initial reference frame image is in luminance-chrominance YUV format, the step of luminance enhancement of the initial reference frame image to obtain the input reference frame image includes: The initial reference frame image is brightened to obtain the enhanced initial reference frame image; The enhanced initial reference frame image is converted to RGB format to obtain the input reference frame image.

7. The method according to claim 6, characterized in that, The step of enhancing the brightness of the initial reference frame image to obtain the enhanced initial reference frame image includes: Obtain the average brightness parameter corresponding to the initial reference frame image; If the brightness quantization index of the initial reference frame image is less than the preset brightness enhancement trigger threshold, the brightness enhancement status flag of the initial reference frame image is set according to the average brightness parameter, and the brightness parameter of the initial reference frame image is shifted to the left to obtain the enhanced initial reference frame image.

8. The method according to claim 1, characterized in that, The initial reference frame image is in YUV format. The process of enhancing the brightness of the initial reference frame image to obtain the input reference frame image includes: The initial reference frame image is converted to RGB format to obtain the reference frame image after the first format conversion. The brightness of the reference frame image after the first format conversion is enhanced to obtain the input reference frame image.

9. The method according to claim 8, characterized in that, The step of enhancing the brightness of the format-converted reference frame image to obtain the input reference frame image includes: Perform YUV conversion on the reference frame image after the first format conversion to obtain the initial reference frame image after the second format conversion; The brightness of the initial reference frame image after the second format conversion is enhanced to obtain the enhanced initial reference frame image. The enhanced initial reference frame image is converted to RGB format to obtain the input reference frame image.

10. The method according to claim 1, characterized in that, The step of performing frame image prediction on the output frame based on the input reference frame image to obtain the initial output frame image includes: The temporal interpolation module is invoked to predict the output frame based on the input reference frame image, thereby obtaining the initial output frame image.

11. The method according to claim 10, characterized in that, The invocation of the temporal interpolation module, which performs frame image prediction on the output frame based on the input reference frame image to obtain the initial output frame image, includes: The temporal interpolation module includes an image interpolation module to generate several candidate frame images; wherein, the output image of the image interpolation module is a single candidate frame image. Select the initial output frame image from the candidate frame images.

12. The method according to claim 11, characterized in that, The method further includes: The candidate frame images output by the image interpolation module are stored in the decoded image buffer and marked for use as temporal reference.

13. The method according to claim 11, characterized in that, The method further includes: The input image of the image interpolation module is used as the reference frame image for the output candidate frame image; Candidate frame images that are not part of the reference frame images to be generated are marked as not to be used for temporal reference; Candidate frame images marked as not used for temporal reference and not as the initial output frame image are removed from the decoded image buffer.

14. The method according to claim 11, characterized in that, The time at which candidate frame images, which serve as the initial output frame images, are output from the decoded image buffer is determined by the time at which the input reference frame images are output from the decoded image buffer, the temporal sampling rate, and the order of the candidate frame images among the input reference frame images.

15. The method according to claim 1, characterized in that, Before the target output frame image is output from the decoded image buffer, a bit depth recovery operation is performed on the target output frame image.

16. A frame image prediction device, characterized in that, The device includes: The enhancement module is used to enhance the brightness of the initial reference frame image to obtain the input reference frame image; The prediction module is used to predict the output frame based on the input reference frame image to obtain an initial output frame image; The restoration module is used to restore the brightness of the initial output frame image to obtain the target output frame image.

17. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 15.

18. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 15.

19. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 15.

Images