Video coding processing method, apparatus and device, storage medium, and product

Abstract

Embodiments of the present application provide a video coding processing method, apparatus and device, a storage medium, and a product. The technical solution provided by the embodiments of the present application comprises: dividing a frame to be processed into a region of interest, a non-region of interest, and a transition region; determining filtering intensity information corresponding to smooth blocks in the non-region of interest, and performing filtering processing on the smooth blocks in said frame on the basis of the filtering intensity information to obtain a filtered image frame; adjusting initial quantization parameters of a plurality of coding units in the filtered image frame determined on the basis of an average bitrate control strategy; and performing coding processing on the filtered image frame on the basis of target quantization parameters of the coding units. In this way, the conditions of blurring and blocking artifact of the non-region of interest due to the allocated low bitrate are reduced while the video coding quality is ensured, so that the technical problems of the reduction of the video coding quality and poor video coding quality easily caused by an average bitrate control strategy under the condition of limited bandwidth are effectively solved, thereby improving the video coding quality.

Description

Video encoding processing method, device, equipment, storage medium and product

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on August 17, 2023, with application number 202311038088.1, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The embodiments of the present application relate to the field of coding technology, and in particular to a video coding processing method, apparatus, device, storage medium, and product. Background Art

[0003] With the development of video coding technology, it has been applied to various fields of life. Different optimization directions have been developed for different application scenarios. In bandwidth-limited scenarios such as entertainment videos and video calls, the area where the human figure is located is more important than the background area, and it is often necessary to enhance the human figure area in video encoding.

[0004] In many application scenarios, low-complexity portrait detection is necessary because of the need to detect people in every frame. Therefore, some low-complexity skin color detection algorithms have been proposed. Their core idea is to convert the current pixel into different color spaces, compare each color component in the color space to see if it matches the empirical value of skin color, and then delineate a skin color region of interest (ROI) containing all skin color pixels. During the encoding process, the bitrate allocated to the skin color region of interest can be increased, thereby improving the subjective quality of this region and thus enhancing the overall quality perception of the human eye.

[0005] However, in bandwidth-limited scenarios like entertainment videos and video calls, the total bitrate available for video encoding is constant, typically based on an average bitrate coding strategy (e.g., ABR, Average Bit Rate). This means the output bitrate must match the input bitrate designed based on bandwidth, ensuring the encoded stream can be sent within the current bandwidth. Under this coding strategy, while increasing the bitrate allocated to areas of interest (such as skin color), it's necessary to reduce the bitrate allocated to areas of non-interest. When bandwidth conditions are suboptimal or the given bitrate is very low, reducing the bitrate in non-interested areas can significantly degrade quality, resulting in subjectively unfavorable blockiness and blurring, ultimately leading to a decrease in video encoding quality and poor quality.

[0006] Summary of the Invention

[0007] The embodiments of the present application provide a video encoding processing method, apparatus, device, storage medium and product to solve the technical problem in the related art that the average bit rate code control strategy in situations with limited bandwidth easily leads to a decline in video encoding quality and poor video encoding quality, thereby effectively improving the video encoding quality.

[0008] In a first aspect, an embodiment of the present application provides a video encoding processing method, comprising:

[0009] Determining a region of interest, a region of no interest, and a transition region in a frame to be processed, wherein the transition region is located between the region of interest and the region of no interest;

[0010] Determining smooth blocks in the non-interested region and filter strength information corresponding to each smooth block, and performing filtering processing on the smooth blocks in the to-be-processed frame according to the filter strength information to obtain a filtered image frame;

[0011] determining initial quantization parameters of a plurality of coding units in the filtered image frame based on a set average bit rate code control strategy, and adjusting the initial quantization parameters corresponding to the region of interest, the region of no interest, and the transition region to obtain a target quantization parameter for each coding unit, wherein the adjustment values ​​of the initial quantization parameters for the region of interest, the transition region, and the region of no interest are increased in sequence;

[0012] The filtered image frame is encoded according to the target quantization parameter of each encoding unit.

[0013] In a second aspect, an embodiment of the present application provides a video encoding processing device, including a region determination module, an image filtering module, a parameter adjustment module, and an image encoding module, wherein:

[0014] The region determination module is configured to determine a region of interest, a region of no interest, and a transition region in a frame to be processed, wherein the transition region is located between the region of interest and the region of no interest;

[0015] The image filtering module is configured to determine smooth blocks in the non-interested region and filtering strength information corresponding to each smooth block, and perform filtering processing on the smooth blocks in the frame to be processed according to the filtering strength information to obtain a filtered image frame;

[0016] The parameter adjustment module is configured to determine initial quantization parameters of a plurality of coding units in the filtered image frame based on a set average bit rate code control strategy, and adjust the initial quantization parameters corresponding to the region of interest, the region of no interest, and the transition region to obtain a target quantization parameter for each coding unit, wherein the adjustment values ​​of the initial quantization parameters of the region of interest, the transition region, and the region of no interest are increased in sequence;

[0017] The image encoding module is configured to perform encoding processing on the filtered image frame according to the target quantization parameter of each of the encoding units.

[0018] In a third aspect, an embodiment of the present application provides a video encoding processing device, including: a memory and one or more processors;

[0019] The memory is used to store one or more programs;

[0020] When the one or more programs are executed by the one or more processors, the one or more processors implement the video encoding processing method as described in the first aspect.

[0021] In a fourth aspect, an embodiment of the present application provides a non-volatile storage medium storing computer-executable instructions, wherein the computer-executable instructions, when executed by a computer processor, are used to perform the video encoding processing method as described in the first aspect.

[0022] In the fifth aspect, an embodiment of the present application provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor of the device reads and executes the computer program from the computer-readable storage medium, so that the device performs the video encoding processing method described in the first aspect.

[0023] The embodiment of the present application divides the frame to be processed into an area of ​​interest, an area of ​​non-interest, and a transition area, determines the filtering strength information corresponding to the smooth block in the area of ​​non-interest, filters the smooth block in the frame to be processed according to the filtering strength information to obtain a filtered image frame, adjusts the initial quantization parameters of multiple coding units in the filtered image frame determined based on the average bit rate code control strategy, and then encodes the filtered image frame according to the target quantization parameter of each coding unit. By filtering the area of ​​non-interest that has less impact on the subjective quality of the human eye, and increasing the adjustment values ​​of the initial quantization parameters of the area of ​​interest, transition area, and area of ​​non-interest in sequence, while ensuring the video encoding quality, the blurring and blocking effects in the area of ​​non-interest due to the low allocated bit rate are reduced, effectively solving the technical problem that the average bit rate code control strategy easily leads to a decline in video encoding quality and poor video encoding quality in situations with limited bandwidth, thereby improving the video encoding quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] FIG1 is a flowchart of a video encoding processing method provided by an embodiment of the present application;

[0025] FIG2 is a flowchart of another video encoding processing method provided by an embodiment of the present application;

[0026] FIG3 is a schematic diagram of an original area division provided in an embodiment of the present application;

[0027] FIG4 is a schematic diagram of a modified area division provided in an embodiment of the present application;

[0028] FIG5 is a schematic diagram showing a display of a region of interest, a region of no interest, and a transition region provided by an embodiment of the present application;

[0029] FIG6 is a structural diagram of a video encoding processing device provided in an embodiment of the present application;

[0030] FIG7 is a schematic structural diagram of a video encoding processing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solutions and advantages of the present application clearer, the specific embodiments of the present application are further described in detail below in conjunction with the accompanying drawings. It is understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. It should also be noted that, for ease of description, only some, but not all, of the contents related to the present application are shown in the accompanying drawings. Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe each operation (or step) as a sequential process, many of the operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of the operations can be rearranged. The above process can be terminated when its operation is completed, but it can also have additional steps not included in the accompanying drawings. The above process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0032] The video coding method provided in this application can be applied to live broadcast scenarios, video call scenarios, or other video transmission scenarios under bitrate / bandwidth constraints. It aims to filter non-interested regions of interest (ROIs) that have little impact on subjective quality, and adjust the initial quantization parameters of multiple coding units in a filtered image frame determined based on an average bitrate coding strategy. The adjustment values ​​for the initial quantization parameters of the ROIs, transition regions, and ROIs are sequentially increased, thereby improving the video coding quality based on the average bitrate coding strategy under limited bandwidth. In existing video coding schemes, when encoding video based on the average bitrate coding strategy under limited bandwidth, the bitrate allocated to the skin color ROI is generally increased while the bitrate allocated to ROIs is reduced. When bandwidth conditions are not ideal or the given bitrate is very low, reducing the bitrate of ROIs can significantly reduce the quality of ROIs, resulting in blocking effects and blurring that are detrimental to the subjective quality of the video, and thus leading to a decrease in overall subjective quality. Based on this, a video encoding processing method according to an embodiment of the present application is provided to solve the technical problem that the average bit rate code control strategy of the existing video encoding scheme easily leads to a decrease in video encoding quality and poor video encoding quality in situations where bandwidth is limited.

[0033] Figure 1 shows a flowchart of a video encoding processing method provided in an embodiment of the present application. The video encoding processing method provided in an embodiment of the present application can be executed by a video encoding processing device, which can be implemented by hardware and / or software and integrated into a video encoding processing device.

[0034] The following description is based on an example of a video encoding processing method executed by a video encoding processing device. Referring to FIG1 , the video encoding processing method includes:

[0035] S110: Determine a region of interest, a region of no interest, and a transition region in the frame to be processed, where the transition region is located between the region of interest and the region of no interest.

[0036] The frame to be processed provided by this solution can be an image frame in the video information to be encoded, awaiting encoding. Exemplarily, after obtaining the frame to be processed that requires encoding, a region of interest (ROI), a non-interest (ROI), and a transition region are determined within the frame to be processed. The transition region serves as a transition between the ROI and the non-interest, and is located between the ROI and the non-interest. The ROI can be understood as the region of interest to the human eye within the frame to be processed. This solution uses the region corresponding to human skin within the frame to be processed as the ROI. During encoding, the encoding bitrate of the ROI is increased, while the encoding bitrate of the non-interest region is reduced. The encoding bitrate of the transition region transitions between the ROI and the non-interest region. The ROI provided by this solution can be the region corresponding to the skin color of a person in the image. For example, the region of interest is formed by multiple pixels whose color values ​​fall within a set color value range for skin color pixels, while the region of non-interest is formed by multiple pixels whose color values ​​do not fall within the set color value range for skin color pixels. The transition region serves as a transition between the ROI and the non-interest region, and its corresponding pixels can be determined from pixels in the non-interest region adjacent to the ROI.

[0037] In one embodiment, the frame to be processed can be divided into multiple image blocks of a set size, each image block includes multiple pixels, and the area corresponding to each image block is determined separately. For example, the area corresponding to the image block can be determined based on the block pixel value corresponding to the image block (for example, the average pixel value of multiple pixels in the image block). The pixel value provided by this solution can be represented by the pixel value of the Y channel reflecting the brightness and the pixel value of the UV channel reflecting the color of the corresponding pixel point or image block in the YUV (brightness, chroma, concentration) color space. Optionally, it can be determined based on the block pixel value of the image block whether the image block belongs to the area of ​​interest (the block pixel value is within the empirical range of the skin color pixel value) or the area of ​​non-interest (the block pixel value is outside the empirical range of the skin color pixel value), and the image blocks adjacent to or close to the area of ​​interest in the area of ​​non-interest can be divided into the transition area.

[0038] Optionally, the size of the image blocks of the frames being processed can be determined based on their resolutions. Since the lower the resolution of the frame being processed or the video information being encoded, the smaller the size, the pixel values ​​corresponding to the central pixel of a set size (e.g., 2x2) can be used to partition the image blocks. This effectively reduces the computational complexity of the video encoding process by dividing the frame being processed into multiple image blocks of the set size. For example, the frame being processed can be divided into NxN image blocks. To ensure the accuracy of skin color recognition, the image blocks can be divided into even smaller blocks for videos with resolutions less than 360p, where N=8. For videos of 360p and above, N=16. Optionally, the pixel values ​​of each NxN image block are recorded (e.g., in a skin_map. Optionally, the skin_map only records the pixel values ​​corresponding to the Y component and does not record the UV component to reduce storage space usage). If the recorded pixel value of the current image block meets the skin color experience value, the current image block is considered to belong to the area of ​​interest; if the recorded pixel value of the current image block does not meet the skin color experience value, the current image block is considered to belong to the area of ​​non-interest, and the image blocks adjacent to the area of ​​interest in the non-interest area are divided into the transition area.

[0039] S120: Determine smooth blocks in the non-interested region and filter strength information corresponding to each smooth block, and perform filtering on the smooth blocks in the frame to be processed according to the filter strength information to obtain a filtered image frame.

[0040] Exemplarily, after determining the region of interest, the region of no interest, and the transition region in the frame to be processed, smooth blocks are determined in each image block of the region of no interest, and filtering strength information corresponding to each smooth block is determined.

[0041] Image blocks in non-regions of interest can be divided into smooth blocks and non-smooth blocks based on the smoothness of their pixel values. Each smooth block corresponds to a filter strength information. When the smoothness of the pixel values ​​in an image block reaches a set threshold, the corresponding image block can be considered a smooth block. When the smoothness of the pixel values ​​in an image block does not reach the set threshold, the corresponding image block can be considered a non-smooth block. Optionally, the higher the smoothness of the pixel values ​​corresponding to the smooth area, the higher the filter strength information can be set.

[0042] In one embodiment, the detection of smooth blocks in non-interested regions can be determined based on the average pixel value of multiple pixels around and in the center of the image block in the non-interested region (the average pixel value can be determined based on a first average pixel value of multiple pixels around and a second average pixel value of multiple pixels in the center) and the difference between the average pixel values ​​of multiple pixels in the center of the image block. The smaller the difference, the higher the corresponding smoothness. The detection of smooth blocks in non-interested regions can also be performed based on edge detection using a Sobel operator. The operator identifies the gradient around each pixel in the image block to determine whether the current pixel is an edge pixel with a significant high frequency. If the proportion of edge pixels in an image block is less than a set proportion threshold, the current image block can be considered a smooth block. In addition, different filtering strength information (e.g., a Gaussian threshold) can also be assigned based on the proportion of edge pixels. Optionally, since the brightness Y component can characterize the texture information of the pixel to the greatest extent, the smoothness of the pixel value can be accurately judged using the texture information. The detection of smooth blocks in non-interested areas can be based on the pixel values ​​of the Y channel reflecting the brightness of multiple pixel points around and in the center of the image block in the non-interested area.

[0043] In one embodiment, for each smooth block, filtering is performed on the smooth block in the frame to be processed according to the filtering strength information to obtain a filtered image frame. The filtering of the smooth block can reduce high-frequency information in the smooth block, thereby reducing blocking artifacts caused by subsequent encoding of the smooth block.

[0044] Optionally, the filtering method for smooth blocks can be Gaussian filtering, mean filtering, median filtering, bilateral filtering (where higher complexity is allowed) or other filtering methods designed to reduce high-frequency information based on video characteristics. For example, in situations with a lot of noise (dark light, unstable acquisition, etc.), the use of median filtering can effectively reduce high-frequency salt and pepper noise in the video, and can also improve the subjective quality of non-skin-color areas (non-areas of interest). By combining skin color detection of the frame to be processed with Gaussian filtering of the non-skin-color areas (non-areas of interest), the skin color detection result is used for the subsequent encoder to adjust the bit rate allocation, and the Gaussian filter will directly process the original image before encoding. By adaptive Gaussian filtering of the non-skin-color areas, the relatively smoother areas are processed, thereby minimizing the loss of high-frequency information and improving the video encoding quality.

[0045] In one embodiment, when filtering an image block in a smooth area according to filtering strength information, each pixel in the image block may be traversed in sequence, and the pixel values ​​of a surrounding area of ​​a set size (a 3x3 or 5x5 area, except at the boundary of the picture, this 3x3 or 5x5 area can obtain pixel points of other image blocks outside this image block) with the corresponding pixel as the center are filtered, and the result is used as the filtered value of the center point.

[0046] S130: Determine initial quantization parameters of multiple coding units in the filtered image frame based on the set average bit rate code control strategy, and adjust the initial quantization parameters corresponding to the region of interest, the region of no interest, and the transition region to obtain a target quantization parameter for each coding unit.

[0047] In this solution, the adjustment values ​​of the initial quantization parameters (QP) for the region of interest, transition region, and non-region of interest are increased in sequence. It should be explained that when encoding an image, after determining the quantization parameters of multiple coding units in the image, each coding unit of the image can be encoded according to the quantization parameters, and the bit rate corresponding to the coding unit after the encoding process is negatively correlated with the quantization parameter, that is, the larger the quantization parameter during the final quantization, the smaller the corresponding bit rate after encoding.

[0048] Exemplarily, the initial quantization parameters of multiple coding units of the filtered image frame obtained by the above filtering processing are determined based on the set average bit rate coding strategy (ABR coding strategy), and the initial quantization parameters corresponding to the area of ​​interest, the area of ​​no interest and the transition area are adjusted respectively to obtain the target quantization parameters of each coding unit.

[0049] During the encoding process, this scheme considers the ROI to be more important and allocates a higher bitrate to it. It considers the transition region to be adjacent to the important region and slightly increases or decreases the bitrate allocated to the transition region. Furthermore, the ROI is considered to be insensitive to the human eye and the bitrate allocated to the ROI is reduced. This scheme adjusts the bitrate allocated to the ROI, ROI, and transition region by adjusting the initial quantization parameters of multiple coding units in the filtered image frame, which is determined based on a set average bitrate control strategy.

[0050] For example, for a region of interest, the initial quantization parameters corresponding to the coding units in the region of interest are reduced (or kept unchanged) to obtain target quantization parameters for the corresponding coding units. When encoding the coding units in the region of interest, a higher bitrate is allocated to these coding units (or kept unchanged). For a non-region of interest, the initial quantization parameters corresponding to the coding units in the non-region of interest are increased (or kept unchanged) to obtain target quantization parameters for the corresponding coding units. When encoding the coding units in the region of interest, a lower bitrate is allocated to these coding units (or kept unchanged). For a transition region, the initial quantization parameters corresponding to the coding units in the region of interest are slightly increased or decreased (or kept unchanged) to obtain target quantization parameters for the corresponding coding units. When encoding the coding units in the region of interest, the bitrate allocated to these coding units is slightly decreased or increased (or kept unchanged) relative to the bitrate allocated according to the initial quantization parameters. The adjusted value of the initial quantization parameter for the transition region is between the adjusted values ​​of the initial quantization parameters for the region of interest and the non-region of interest.

[0051] In one possible embodiment, the present solution provides a first adjustment value for adjusting the initial quantization parameter corresponding to the region of interest, which is less than or equal to 0; a second adjustment value for adjusting the initial quantization parameter corresponding to the non-region of interest, which is greater than or equal to 0; and a third adjustment value for adjusting the initial quantization parameter corresponding to the transition region, which is between the first adjustment value and the second adjustment value. When adjusting the initial quantization parameters corresponding to the region of interest, the non-region of interest, and the transition region, the sum of the initial quantization parameter and the corresponding adjustment value may be used as the target quantization parameter. The third adjustment values ​​corresponding to different coding units may be the same or different. This solution adjusts the initial quantization parameters corresponding to the region of interest, the non-region of interest, and the transition region using the first adjustment value, the second adjustment value, and the third adjustment value, respectively, to accurately obtain the target quantization parameters corresponding to the different regions. Subsequently, encoding the filtered image frame to obtain an encoded image maintains the display quality of skin color in the region of interest while providing smoother and more natural transitions between different regions. This ensures video encoding quality while reducing blur and blocking artifacts in the non-region of interest due to an excessively low allocated bitrate.

[0052] In one embodiment, when multiple pixels of a coding unit are located in different regions (including a region of interest, a region of no interest, and a transition region), an initial quantization parameter of the coding unit may be adjusted based on adjustment values ​​corresponding to the different regions (including a first adjustment value, a second adjustment value, and a third adjustment value) to obtain a target quantization parameter for the coding unit. For example, an average of the adjustment values ​​of the regions where the multiple pixels of the coding unit are located may be used as the adjustment value for adjusting the initial quantization parameter of the coding unit, or a weighted summation operation may be performed on multiple adjustment values ​​based on the proportion of pixels in different regions, and the weighted summation operation result may be used as the adjustment value for adjusting the initial quantization parameter of the coding unit.

[0053] S140: Encoding the filtered image frame according to the target quantization parameter of each coding unit.

[0054] Exemplarily, when determining the target quantization parameter corresponding to each coding unit, the filtered image frame is encoded according to the target quantization parameter of each coding unit to obtain an encoded image. Since the target quantization parameter of the region of interest, the region of no interest, and the transition region may change relative to the initial quantization parameter, for the coding unit where the quantization parameter changes, when the target quantization parameter increases relative to the initial quantization parameter, the bit rate allocated during the encoding process decreases, and when the target quantization parameter decreases relative to the initial quantization parameter, the bit rate allocated during the encoding process increases.

[0055] Among them, in the final encoded image, the total bit rate corresponding to the region of interest is improved compared with the total bit rate corresponding to the encoded image obtained by encoding based on the initial quantization parameters, the total bit rate corresponding to the non-interest region is reduced compared with the total bit rate corresponding to the encoded image obtained by encoding based on the initial quantization parameters, and the total bit rate corresponding to the transition region is slightly reduced or increased compared with the total bit rate corresponding to the transition region of the encoded image obtained by encoding based on the initial quantization parameters. The display quality in the region of interest is higher, and the display in the transition region is smoother and more natural, which effectively reduces the blur and block effects in the non-interest region due to the low allocated bit rate while ensuring the video encoding quality.

[0056] In one embodiment, when encoding the filtered image frame according to the target quantization parameter of each coding unit, the target quantization parameter is first rounded off and then converted to an integer-based target quantization parameter.

[0057] In the above, by dividing the frame to be processed into an area of ​​interest, an area of ​​non-interest and a transition area, determining the filtering strength information corresponding to the smooth block in the area of ​​non-interest, filtering the smooth block in the frame to be processed according to the filtering strength information to obtain a filtered image frame, and adjusting the initial quantization parameters of multiple coding units in the filtered image frame determined based on the average bit rate coding strategy, and then encoding the filtered image frame according to the target quantization parameter of each coding unit, filtering the area of ​​non-interest that has less impact on the subjective quality of the human eye, and increasing the adjustment values ​​of the initial quantization parameters of the area of ​​interest, transition area and area of ​​non-interest in sequence, while ensuring the video encoding quality, reducing the blurring and blocking effects in the area of ​​non-interest due to the low allocated bit rate, effectively solving the technical problem that the average bit rate coding strategy easily leads to a decrease in video encoding quality and poor video encoding quality in situations with limited bandwidth, thereby improving the video encoding quality.

[0058] Based on the above embodiment, FIG2 shows a flowchart of another video encoding processing method provided by an embodiment of the present application. The video encoding processing method is a specific embodiment of the above video encoding processing method. Referring to FIG2, the video encoding processing method includes:

[0059] S210: Determine a first pixel value of each image block in a frame to be processed.

[0060] S220: Divide each image block into a region of interest or a region of no interest based on the first pixel value and a set skin color experience range.

[0061] S230: Divide the image blocks adjacent to the region of interest in the non-region of interest into a transition region.

[0062] Exemplarily, the frame to be processed is divided into a plurality of image blocks of a set size (e.g., 8x8), and the first pixel value of each image block is calculated. Optionally, the first pixel value can be represented by the average pixel value of a plurality of pixels within a set range (e.g., 2x2) in the image block.

[0063] The first pixel value of each image block is compared with a set skin color experience range (the pixel value range corresponding to skin color). Image blocks with first pixel values ​​within the skin color experience range are classified as regions of interest, while image blocks with first pixel values ​​outside the skin color experience range are classified as regions of no interest. After determining whether each image block belongs to a region of interest or a region of no interest, image blocks in the region of no interest that are adjacent to the region of interest are classified as transition regions. In this case, the transition region is located between the region of interest and the region of no interest.

[0064] This solution divides the frame to be processed into the region of interest or the region of non-interest by comparing the first pixel value of each image block with the skin color experience range, and divides the image blocks adjacent to the region of interest in the non-interest area into the transition area, accurately dividing the frame to be processed into the region of interest, the region of non-interest and the transition area, and making the subsequent filtering processing of the region of non-interest more accurate, thereby improving the video encoding quality.

[0065] In one possible embodiment, the video encoding processing method provided by the present solution further includes at least one of S231 and S232 after dividing each image block into an area of ​​interest or an area of ​​non-interest based on the first pixel value and the set skin color experience range (and before dividing image blocks adjacent to the area of ​​interest in the area of ​​non-interest into the transition area):

[0066] S231: Converting an independent image block in the region of interest to a non-region of interest.

[0067] S232: Convert image blocks in the non-region of interest whose number of adjacent image blocks to the region of interest reaches a set number into the region of interest.

[0068] In one embodiment, after each image block is divided into a region of interest or a region of no interest, an independent image block is determined within the region of interest and the independent image block is transferred from the region of interest to the region of no interest. Optionally, if the number of the eight image blocks surrounding the image block within the region of interest that belongs to the region of interest is less than or equal to a predetermined first number (e.g., 1-2), the corresponding image block may be considered an independent image block.

[0069] In one embodiment, after each image block is divided into a region of interest or a non-region of interest, image blocks in the non-region of interest that are adjacent to the region of interest are identified, and the number of adjacent blocks between these image blocks and the region of interest is determined (i.e., the number of adjacent blocks in the eight image blocks surrounding the image block that belong to the region of interest). Image blocks in the non-region of interest whose corresponding number of adjacent blocks reaches a set number are converted to the region of interest. Optionally, when the number of adjacent blocks reaches a set second number (e.g., 6 to 8), the image block in the non-region of interest can be considered an image block surrounded by the region of interest and converted to the region of interest.

[0070] This solution converts independent image blocks in the region of interest into the non-region of interest, and converts image blocks in the non-region of interest that are surrounded by the region of interest into the region of interest, thereby reducing the situation where an independent region without a bit rate or an independent region with a very high bit rate is allocated when the bit rate is subsequently allocated to image blocks in different regions, reducing the misjudgment of skin color areas, and improving video encoding quality.

[0071] For example, after dividing each image block into an area of ​​interest or an area of ​​non-interest based on the first pixel value and the set skin color experience range, and dividing the image blocks adjacent to the area of ​​interest in the area of ​​non-interest into the transition area, an original area division schematic diagram as shown in Figure 3 is obtained, in which each small grid is an image block, where area A is the area of ​​interest and area B is the area of ​​non-interest, and different areas are distinguished by different textures.

[0072] Assume that the first number is set to 2 and the second number is set to 6. Among the eight image blocks surrounding image blocks a, b, and c in the region of interest, the number of image blocks belonging to the region of interest is all 1, which is less than the set first number. Image blocks a, b, and c in the region of interest can be considered as independent image blocks in the region of interest, and image blocks a, b, and c can be divided into the non-region of interest. Image blocks d, e, and f in the non-region of interest have 6 adjacent regions to the region of interest, which reaches the set second number. Image blocks d, e, and f can be converted into the region of interest. At this point, a revised region division diagram is obtained as shown in Figure 4. Based on Figure 4, the image blocks adjacent to the region of interest in the non-region of interest are divided into transition regions, resulting in a display diagram of the region of interest, non-region of interest, and transition regions as shown in Figure 5, where region A is the region of interest, region B is the non-region of interest, and region C is the transition region.

[0073] S240: Determine a pixel change value of each image block in the non-region of interest.

[0074] S250: determining smooth blocks from a plurality of image blocks in a non-interested region according to a smoothing threshold range corresponding to the pixel change value, and determining filtering strength information corresponding to each smooth block.

[0075] Wherein, different smoothing threshold ranges correspond to different filtering strength information. Exemplarily, the pixel change value of each image block in the non-interested area is calculated, and the smoothing threshold range corresponding to the pixel change value of each image block is determined. The smoothed block is determined in multiple image blocks in the non-interested area according to the smoothing threshold range corresponding to the pixel change value, and the filtering strength information corresponding to the smoothed block is determined according to the pre-set filtering strength information corresponding to different smoothed blocks. Wherein, the smaller the value corresponding to the smoothing threshold range, the greater the corresponding filtering strength information. This solution accurately determines the smoothed block according to the smoothing threshold range corresponding to the pixel change value of each image block in the non-interested area, and determines the filtering strength information corresponding to each smoothed block according to the smoothing threshold range, thereby improving the efficiency of determining the filtering strength information, reducing the filtering processing of non-smooth blocks, and ensuring the quality of video encoding.

[0076] In one embodiment, the video encoding processing method provided by the present solution includes, when determining the pixel change value of each image block in the non-interested region:

[0077] S241: Determine a first average pixel value of each image block in a non-interest region in a first pixel range and a second average pixel value in a second pixel range.

[0078] S242: Determine a pixel change value of each image block in the non-interest region according to the first average pixel value and the second average pixel value.

[0079] Exemplarily, a first average pixel value of a first pixel range and a second average pixel value of a second pixel range are determined for each image block in the non-region of interest, where different pixels exist in the first pixel range and the second pixel range. For example, the first pixel range may be a pixel range consisting of pixels at the four corners of the image block, and the second pixel range may be a pixel range consisting of pixels of a set size (e.g., 2x2) at the center of the image block. Optionally, the second average pixel value may be the first pixel value of the corresponding image block in the frame to be processed determined above.

[0080] The pixel change value for each image block in the non-interest region is determined based on the first average pixel value and the second average pixel value determined above. For example, for an image block, the absolute value of the difference between the corresponding first average pixel value and the second average pixel value can be used as the pixel change value of the image block. For example, assuming that the pixel values ​​at the upper left corner, upper right corner, lower left corner, and lower right corner of the image block are x_tl, x_tr, x_bl, and x_br, respectively, then the first average pixel value is avg(x_tl+x_tr+x_bl+x_br). x is the sequence number of the image block, and the second average pixel value can be represented by the first pixel value skin_map[x] of the corresponding image block in the frame to be processed determined above. Then, the pixel change value of the image block can be expressed as abs(avg(x_tl+x_tr+x_bl+x_br)-skin_map[x]). This solution accurately determines the pixel change value of each image block in the non-interest area based on the first average pixel value of each image block in the first pixel range and the second average pixel value in the second pixel range of each image block in the non-interest area. While reducing the amount of calculation of the pixel change value, it correctly reflects the degree of change of the pixel points in the image block, accurately distinguishes smooth blocks and non-smooth blocks, and improves the video encoding quality.

[0081] In one possible embodiment, the video encoding processing method provided by the present solution includes determining smooth blocks from multiple image blocks in a non-interested region based on a smoothing threshold range corresponding to a pixel change value, and determining filtering strength information corresponding to each smooth block, including:

[0082] When the pixel change value of the image block in the non-interested region is within the first smoothing threshold range, determining the image block as a first smoothing block, and determining the filtering strength information corresponding to the first smoothing block as the first filtering strength information;

[0083] When a pixel change value of the image block in the non-region of interest is within a second smoothing threshold range, determining the image block as a second smoothing block, and determining the filtering strength information corresponding to the second smoothing block as second filtering strength information, the second smoothing threshold range is higher than the first smoothing threshold range, and the first filtering strength information is greater than the second filtering strength information;

[0084] When the pixel change value of the image block in the non-region of interest is outside the second smoothing threshold range, the image block is determined to be a non-smooth block, and the filtering strength information corresponding to the non-smooth block is determined to be the third filtering strength information, the second smoothing threshold range is higher than the first smoothing threshold range, and the second filtering strength information is greater than the third filtering strength information.

[0085] Exemplarily, for each image block in a non-region of interest, a smoothing threshold range corresponding to the pixel change value of the image block is determined. When the pixel change value is within the first smoothing threshold range, the image block is determined to be a first smooth block, and the filtering strength information corresponding to the first smooth block is determined to be the first filtering strength information. When the pixel change value is within the second smoothing threshold range, the image block is determined to be a second smooth block, and the filtering strength information corresponding to the second smooth block is determined to be the second filtering strength information. If the pixel change value is outside the second smoothing threshold range, the image block is determined to be a non-smooth block, and the filtering strength information corresponding to the non-smooth block is determined to be the third filtering strength information. The smoothness level corresponding to the first smooth block is greater than the smoothness level of the second smooth block.

[0086] Optionally, the filtering strength corresponding to the third filtering strength information may be set to 0, that is, no filtering process is performed on the image block corresponding to the third filtering strength information.

[0087] For example, the filter strength information may be determined based on the following formula:

[0088] Where th1 is the lower limit of the first smoothing threshold range, th2 is the upper limit of the second smoothing threshold range, and flat_a, flat_b, and flat_c are regions indicating different degrees of smoothness, with the smoothness corresponding to regions flat_a, flat_b, and flat_c decreasing in order. Optionally, different filter strength information can be pre-configured for regions flat_a, flat_b, and flat_c. For example, the filter strength information corresponding to region flat_a is the first filter strength information, the filter strength information corresponding to region flat_b is the second filter strength information, and the filter strength information corresponding to region flat_c is the third filter strength information, with the first filter strength information being greater than the second filter strength information, and the second filter strength information being greater than the third filter strength information. Optionally, the third filter strength information can be 0, meaning that when the filter strength information is the third filter strength information, no filtering is performed on the region corresponding to the third filter strength information. This solution improves video encoding quality by accurately distinguishing between smooth and non-smooth blocks and the smoothness of smooth blocks based on the smoothing threshold range corresponding to the pixel change value of the image block, and accurately determining the filter strength information.

[0089] In one possible embodiment, the video encoding processing method provided by this solution further includes: determining smooth blocks from multiple image blocks in a non-interested area according to a smoothing threshold range corresponding to a pixel change value, and determining filtering strength information corresponding to each smooth block.

[0090] When a pixel change value of the image block in the non-region of interest is within a first smoothing threshold range and the image block is adjacent to a transition region, determining the image block as a second smoothing block, and determining the filtering strength information corresponding to the second smoothing block as the second filtering strength information;

[0091] When the pixel change value of the image block in the non-region of interest is within the second smoothing threshold range and the image block is adjacent to the transition region, the image block is determined to be a non-smooth block, and the filtering strength information corresponding to the non-smooth block is determined to be the third filtering strength information.

[0092] Exemplarily, when a pixel change value of an image block in a non-region of interest is within a first smoothing threshold range and the image block is adjacent to a transition region, the image block is determined to be a second smoothing block, and the filter strength information corresponding to the second smoothing block is determined to be the second filter strength information. Furthermore, when a pixel change value of an image block in a non-region of interest is within the first smoothing threshold range but the image block is not adjacent to a transition region, the corresponding image block is determined to be a first smoothing block.

[0093] When the pixel change value of the image block in the non-interested region is within the second smoothing threshold range and the image block is adjacent to the transition region, the image block is determined to be a non-smooth block, and the filter strength information corresponding to the non-smooth block is determined to be the third filter strength information. When the pixel change value of the image block in the non-interested region is within the second smoothing threshold range but the image block is not adjacent to the transition region, the corresponding image block is the second smooth block. Since areas near skin color are also relatively easy for the human eye to perceive, such as human hair, clothing, handheld objects, etc., in order to avoid the negative impact of excessive filter strength on encoding quality in these areas, this solution downgrades the filter strength information of smooth blocks adjacent to transition regions, reduces the filter strength of adjacent transition regions, and improves video encoding quality.

[0094] In one possible embodiment, the video encoding processing method provided by this solution further includes: performing filter strength transition processing on the filter strength information corresponding to the smooth blocks adjacent to the non-filtered blocks after determining the smooth blocks in the non-interest area and the filter strength information corresponding to each smooth block.

[0095] Among them, the non-filtered blocks can be image blocks in the transition area and image blocks in the non-interested area that are considered to be non-smooth blocks. These blocks are all image blocks that do not need to be filtered. Exemplarily, for smooth blocks adjacent to non-filtered blocks, the filter strength information of these smooth blocks is subjected to filter strength transition processing. For example, for a smooth block adjacent to a non-filtered block, the closer it is to the pixel point of the adjacent non-filtered area, the lower the filter strength corresponding to its corresponding filter strength information. At this time, there are multiple types of filter strength information corresponding to different pixel points in the smooth area.

[0096] Optionally, when performing filter strength transition processing on the filter strength information corresponding to the smooth block adjacent to the non-filtered block, the filter strength transition processing can be performed on a block smaller than the smooth block size (for example, a 4x4 block). If the filter strengths of two adjacent smooth blocks are different, then if there is no transition between the two smooth blocks, for example, two adjacent smooth blocks A and B, smooth block A is strongly filtered (filtered based on the first filter strength information) and smooth block B is not filtered, the timing coding effect is actually easy to see that there is a clear boundary between blocks AB. This solution performs a relatively narrow filter strength transition processing between smooth blocks A and B, gradually reducing the filter strength in this transition area, rather than transitioning the smooth block of the original size (8x8 or 16x16). It does not need to directly adjust the filter strength of the entire smooth block, further improving the video encoding quality.

[0097] In one possible embodiment, the video encoding processing method provided by this solution further includes: performing filter strength transition processing on the filter strength information corresponding to blocks with different and adjacent filter strength information after determining the smooth blocks in the non-interest area and the filter strength information corresponding to each smooth block.

[0098] Exemplarily, filter strength transition processing is performed on the filter strength information corresponding to adjacent blocks with different filter strength information (e.g., a first smooth block and a second smooth block). For example, the filter strength information corresponding to a set range of adjacent boundaries (e.g., a 2xN range) within the adjacent blocks with different filter strength information is set to the average of the two filter strength information. By performing filter strength transition processing on the filter strength information corresponding to adjacent blocks with different filter strength information, this solution effectively reduces the situation where the filter strength differences between adjacent smooth blocks create significant contrast between their boundaries, thereby ensuring video encoding quality.

[0099] It's important to explain that during image encoding, encoding high-frequency information consumes some bitrate. This process converts image information from the time domain to the frequency domain, ultimately encoding the coefficients of the image's frequency domain information. Smoother regions in the image have more concentrated frequency domain information, resulting in lower bitrates for those regions. In situations where bitrate or bandwidth are limited, this solution performs smooth transition processing on non-interested regions, allowing regions of interest and transitional regions to receive more bitrate, effectively improving video encoding quality. Subsequent encoding operations reduce the bitrate allocated to non-interested regions to increase the bitrate allocated to regions of interest and transitional regions. However, if the bitrate allocation is low and non-interested regions still contain a high amount of high-frequency information, blocking artifacts may occur, and relatively smooth regions are more susceptible to significant blocking artifacts. This solution performs filter strength transition processing on filter strength information in non-interested regions, effectively reducing blocking artifacts in non-interested regions and improving video encoding quality.

[0100] S260: Filter the smooth block in the frame to be processed according to the filter strength information to obtain a filtered image frame.

[0101] S270: Determine initial quantization parameters of multiple coding units in the filtered image frame based on the set average bit rate code control strategy, and adjust the initial quantization parameters corresponding to the region of interest, the region of no interest, and the transition region to obtain a target quantization parameter for each coding unit.

[0102] The adjustment values ​​of the initial quantization parameters of the region of interest, the transition region, and the region of no interest are increased in sequence.

[0103] S280: Encoding the filtered image frame according to the target quantization parameter of each coding unit.

[0104] In the above, by dividing the frame to be processed into an area of ​​interest, an area of ​​non-interest and a transition area, determining the filtering strength information corresponding to the smooth block in the area of ​​non-interest, filtering the smooth block in the frame to be processed according to the filtering strength information to obtain a filtered image frame, and adjusting the initial quantization parameters of multiple coding units in the filtered image frame determined based on the average bit rate coding strategy, and then encoding the filtered image frame according to the target quantization parameter of each coding unit, filtering the area of ​​non-interest that has less impact on the subjective quality of the human eye, and increasing the adjustment values ​​of the initial quantization parameters of the area of ​​interest, transition area and area of ​​non-interest in sequence, while ensuring the video encoding quality, reducing the blurring and blocking effects in the area of ​​non-interest due to the low allocated bit rate, effectively solving the technical problem that the average bit rate coding strategy easily leads to a decrease in video encoding quality and poor video encoding quality in situations with limited bandwidth, thereby improving the video encoding quality. At the same time, by comparing the first pixel value of each image block with the empirical skin color range, the frame to be processed is divided into a region of interest or a region of no interest. Image blocks adjacent to the region of interest in the region of no interest are also divided into transition regions. This accurately divides the frame to be processed into regions of interest, no interest, and transition regions, making subsequent filtering of the no interest regions more accurate and improving video encoding quality. Furthermore, by accurately determining smoothed blocks based on the smoothing threshold range corresponding to the pixel change value of each image block in the region of no interest, and determining the filter strength information corresponding to each smoothed block based on the smoothing threshold range, the efficiency of determining filter strength information is improved, and filtering of non-smoothed blocks is reduced, thus ensuring video encoding quality.

[0105] FIG6 is a schematic diagram of the structure of a video encoding processing device provided by an embodiment of the present application. Referring to FIG6 , the video encoding processing device includes a region determination module 61 , an image filtering module 62 , a parameter adjustment module 63 and an image encoding module 64 .

[0106] Among them, the area determination module 61 is configured to determine the area of ​​interest, the area of ​​non-interest and the transition area in the frame to be processed, and the transition area is located between the area of ​​interest and the area of ​​non-interest; the image filtering module 62 is configured to determine the smooth blocks in the area of ​​non-interest and the filtering strength information corresponding to each smooth block, and filter the smooth blocks in the frame to be processed according to the filtering strength information to obtain a filtered image frame; the parameter adjustment module 63 is configured to determine the initial quantization parameters of multiple coding units in the filtered image frame based on the set average bit rate code control strategy, and adjust the initial quantization parameters corresponding to the area of ​​interest, the area of ​​non-interest and the transition area to obtain the target quantization parameter of each coding unit, wherein the adjustment values ​​of the initial quantization parameters of the area of ​​interest, the transition area and the area of ​​non-interest increase in sequence; the image encoding module 64 is configured to encode the filtered image frame according to the target quantization parameter of each encoding unit.

[0107] In the above, by dividing the frame to be processed into an area of ​​interest, an area of ​​non-interest and a transition area, determining the filtering strength information corresponding to the smooth block in the area of ​​non-interest, filtering the smooth block in the frame to be processed according to the filtering strength information to obtain a filtered image frame, and adjusting the initial quantization parameters of multiple coding units in the filtered image frame determined based on the average bit rate coding strategy, and then encoding the filtered image frame according to the target quantization parameter of each coding unit, filtering the area of ​​non-interest that has less impact on the subjective quality of the human eye, and increasing the adjustment values ​​of the initial quantization parameters of the area of ​​interest, transition area and area of ​​non-interest in sequence, while ensuring the video encoding quality, reducing the blurring and blocking effects in the area of ​​non-interest due to the low allocated bit rate, effectively solving the technical problem that the average bit rate coding strategy easily leads to a decrease in video encoding quality and poor video encoding quality in situations with limited bandwidth, thereby improving the video encoding quality.

[0108] In a possible embodiment, the region determination module 61 is specifically configured as follows:

[0109] Determining a first pixel value of each image block in the frame to be processed;

[0110] Based on the first pixel value and a set skin color experience range, each image block is divided into an area of ​​interest or an area of ​​no interest;

[0111] The image blocks adjacent to the region of interest in the non-region of interest are divided into the transition region.

[0112] In one possible embodiment, the video encoding processing device also includes a region correction module, which is configured to: convert independent image blocks in the region of interest to the non-region of interest; and / or convert image blocks in the non-region of interest whose number of adjacent blocks to the region of interest reaches a set number to the region of interest.

[0113] In a possible embodiment, when determining the smooth blocks in the non-interest region and the filter strength information corresponding to each smooth block, the parameter adjustment module 63 is configured as follows:

[0114] determining a pixel change value for each image block in a non-region of interest;

[0115] Smooth blocks are determined in multiple image blocks of the non-interested area according to the smoothing threshold range corresponding to the pixel change value, and filtering strength information corresponding to each smooth block is determined, wherein different smoothing threshold ranges correspond to different filtering strength information.

[0116] In a possible embodiment, when determining the pixel change value of each image block in the non-interest region, the parameter adjustment module 63 is configured as follows:

[0117] Determine a first average pixel value of a first pixel range and a second average pixel value of a second pixel range for each image block in the non-region of interest;

[0118] A pixel change value of each image block in the non-interested region is determined according to the first average pixel value and the second average pixel value.

[0119] In one possible embodiment, when the parameter adjustment module 63 determines a smooth block from a plurality of image blocks in a non-region of interest according to a smoothing threshold range corresponding to a pixel change value, and determines the filtering strength information corresponding to each smooth block, the configuration is as follows:

[0120] When the pixel change value of the image block in the non-interested region is within the first smoothing threshold range, determining the image block as a first smoothing block, and determining the filtering strength information corresponding to the first smoothing block as the first filtering strength information;

[0121] When a pixel change value of the image block in the non-region of interest is within a second smoothing threshold range, determining the image block as a second smoothing block, and determining the filtering strength information corresponding to the second smoothing block as second filtering strength information, the second smoothing threshold range is higher than the first smoothing threshold range, and the first filtering strength information is greater than the second filtering strength information;

[0122] When the pixel change value of the image block in the non-region of interest is outside the second smoothing threshold range, the image block is determined to be a non-smooth block, and the filtering strength information corresponding to the non-smooth block is determined to be the third filtering strength information, the second smoothing threshold range is higher than the first smoothing threshold range, and the second filtering strength information is greater than the third filtering strength information.

[0123] In a possible embodiment, the parameter adjustment module 63 is further configured to:

[0124] When a pixel change value of the image block in the non-region of interest is within a first smoothing threshold range and the image block is adjacent to a transition region, determining the image block as a second smoothing block, and determining the filtering strength information corresponding to the second smoothing block as the second filtering strength information;

[0125] When the pixel change value of the image block in the non-region of interest is within the second smoothing threshold range and the image block is adjacent to the transition region, the image block is determined to be a non-smooth block, and the filtering strength information corresponding to the non-smooth block is determined to be the third filtering strength information.

[0126] In a possible embodiment, the video encoding processing apparatus further includes a first transition module configured to perform filter strength transition processing on filter strength information corresponding to a smooth block adjacent to a non-filter block.

[0127] In a possible embodiment, the video encoding processing apparatus further includes a second transition module configured to perform filter strength transition processing on filter strength information corresponding to adjacent blocks having different filter strength information.

[0128] In a possible embodiment, the first adjustment value for adjusting the initial quantization parameter corresponding to the region of interest is less than or equal to 0, the second adjustment value for adjusting the initial quantization parameter corresponding to the non-region of interest is greater than or equal to 0, and the third adjustment value for adjusting the initial quantization parameter corresponding to the transition region is between the first adjustment value and the second adjustment value.

[0129] It is worth noting that in the embodiment of the above-mentioned video encoding processing device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the protection scope of the embodiments of this application.

[0130] The embodiment of the present application also provides a video encoding processing device, which can integrate the video encoding processing apparatus provided by the embodiment of the present application. Figure 7 is a structural diagram of a video encoding processing device provided by the embodiment of the present application. Referring to Figure 7, the video encoding processing device includes: an input device 73, an output device 74, a memory 72 and one or more processors 71; the memory 72 is used to store one or more programs; when one or more programs are executed by one or more processors 71, the one or more processors 71 implement the video encoding processing method provided by the above embodiment. The video encoding processing apparatus, device and computer provided above can be used to execute the video encoding processing method provided by any of the above embodiments, and have corresponding functions and beneficial effects.

[0131] The embodiment of the present application also provides a non-volatile storage medium that stores computer-executable instructions, and the computer-executable instructions are used to execute the video encoding processing method provided in the above embodiment when executed by a computer processor. Of course, the non-volatile storage medium that stores computer-executable instructions provided in the embodiment of the present application, its computer-executable instructions are not limited to the video encoding processing method provided above, and can also execute related operations in the video encoding processing method provided in any embodiment of the present application. The video encoding processing device, equipment and storage medium provided in the above embodiment can execute the video encoding processing method provided in any embodiment of the present application. For technical details not described in detail in the above embodiment, please refer to the video encoding processing method provided in any embodiment of the present application.

[0132] On the basis of the above embodiments, the embodiments of the present application also provide a computer program product. The essence of the technical solution of the present application or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer program product is stored in a storage medium and includes a number of instructions for enabling a computer device, a mobile terminal or a processor therein to execute all or part of the steps of the video encoding processing method provided in each embodiment of the present application.

Claims

1. A video encoding processing method, wherein: include: Determine an area of ​​interest, an area of ​​no interest, and a transition area in a frame to be processed, wherein the transition area is located between the area of ​​interest and the area of ​​no interest; Determine the smooth blocks in the non-interested region and the filter strength information corresponding to each of the smooth blocks, and perform filtering processing on the smooth blocks in the frame to be processed according to the filter strength information to obtain a filtered image frame; Determining initial quantization parameters of a plurality of coding units in the filtered image frame based on a set average bit rate code control strategy, and adjusting the initial quantization parameters corresponding to the region of interest, the region of no interest, and the transition region to obtain a target quantization parameter of each coding unit, wherein the adjustment values ​​of the initial quantization parameters of the region of interest, the transition region, and the region of no interest are increased in sequence; The filtered image frame is encoded according to the target quantization parameter of each encoding unit.

2. The video encoding processing method according to claim 1, wherein: The determining of the region of interest, the region of non-interest, and the transition region in the frame to be processed includes: Determining a first pixel value of each image block in a frame to be processed; Based on the first pixel value and a set skin color experience range, each of the image blocks is divided into an area of ​​interest or an area of ​​no interest; The image block adjacent to the region of interest in the non-region of interest is divided into a transition region.

3. The video encoding processing method according to claim 2, wherein: After dividing each of the image blocks into an area of ​​interest or an area of ​​no interest based on the first pixel value and the set skin color experience range, the method further includes at least one of the following: Converting the image block existing independently in the region of interest to a non-region of interest; The image blocks in the non-interest region whose number of adjacent blocks to the interest region reaches a set number are converted to the interest region.

4. The video encoding processing method according to claim 1, wherein: The determining of the smooth blocks in the non-interested region and the filtering strength information corresponding to each of the smooth blocks includes: Determining a pixel change value of each image block in the non-interested region; A smooth block is determined in the plurality of image blocks in the non-interested area according to a smoothing threshold range corresponding to the pixel change value, and filtering strength information corresponding to each smooth block is determined, wherein different smoothing threshold ranges correspond to different filtering strength information.

5. The video encoding processing method according to claim 4, wherein: The determining of the pixel change value of each image block in the non-interested region comprises: Determine a first average pixel value in a first pixel range and a second average pixel value in a second pixel range for each image block in the non-interested region; A pixel change value of each image block in the non-interested region is determined according to the first average pixel value and the second average pixel value.

6. The video encoding processing method according to claim 4, wherein: The step of determining a smooth block from a plurality of the image blocks in the non-interested region according to a smoothing threshold range corresponding to the pixel change value, and determining filtering strength information corresponding to each of the smooth blocks, comprises: When the pixel change value of the image block in the non-interested region is within a first smoothing threshold range, determining the image block as a first smoothing block, and determining the filtering strength information corresponding to the first smoothing block as the first filtering strength information; In a case where the pixel change value of the image block in the non-interested region is within a second smoothing threshold range, determining that the image block is a second smoothing block, and determining that the filtering strength information corresponding to the second smoothing block is second filtering strength information, the second smoothing threshold range is higher than the first smoothing threshold range, and the first filtering strength information is greater than the second filtering strength information; When the pixel change value of the image block in the non-interest area is outside the second smoothing threshold range, the image block is determined to be a non-smooth block, and the filtering strength information corresponding to the non-smooth block is determined to be the third filtering strength information, the second smoothing threshold range is higher than the first smoothing threshold range, and the second filtering strength information is greater than the third filtering strength information.

7. The video encoding processing method according to claim 4, wherein: The step of determining a smooth block from a plurality of the image blocks in the non-interested region according to a smoothing threshold range corresponding to the pixel change value, and determining filtering strength information corresponding to each of the smooth blocks, further includes: In a case where the pixel change value of the image block in the non-interested region is within a first smoothing threshold range and the image block is adjacent to the transition region, determining that the image block is a second smoothing block, and determining that the filtering strength information corresponding to the second smoothing block is the second filtering strength information; When the pixel change value of the image block in the non-interest area is within the second smoothing threshold range and the image block is adjacent to the transition area, the image block is determined to be a non-smooth block, and the filtering strength information corresponding to the non-smooth block is determined to be the third filtering strength information.

8. The video encoding processing method according to claim 1, wherein: After determining the smooth blocks in the non-interested region and the filtering strength information corresponding to each of the smooth blocks, the method further includes: The filter strength information corresponding to the smoothing block adjacent to the non-filtering block is subjected to filter strength transition processing.

9. The video encoding processing method according to claim 1, wherein: After determining the smooth blocks in the non-interested region and the filtering strength information corresponding to each of the smooth blocks, the method further includes: The filter strength information corresponding to the adjacent blocks with different filter strength information is subjected to filter strength transition processing.

10. The video encoding processing method according to any one of claims 1 to 9, wherein: The first adjustment value for adjusting the initial quantization parameter corresponding to the region of interest is less than or equal to 0, the second adjustment value for adjusting the initial quantization parameter corresponding to the non-region of interest is greater than or equal to 0, and the third adjustment value for adjusting the initial quantization parameter corresponding to the transition region is between the first adjustment value and the second adjustment value.

11. A video encoding processing device, wherein: It includes a region determination module, an image filtering module, a parameter adjustment module and an image encoding module, wherein: The region determination module is configured to determine an area of ​​interest, an area of ​​no interest, and a transition area in the frame to be processed, wherein the transition area is located between the area of ​​interest and the area of ​​no interest; The image filtering module is configured to determine the smooth blocks in the non-interested region and the filtering strength information corresponding to each of the smooth blocks, and perform filtering processing on the smooth blocks in the frame to be processed according to the filtering strength information to obtain a filtered image frame; The parameter adjustment module is configured to determine initial quantization parameters of multiple coding units in the filtered image frame based on a set average bit rate code control strategy, and adjust the initial quantization parameters corresponding to the region of interest, the region of no interest, and the transition region to obtain a target quantization parameter of each coding unit, wherein the adjustment values ​​of the initial quantization parameters of the region of interest, the transition region, and the region of no interest are increased in sequence; The image encoding module is configured to perform encoding processing on the filtered image frame according to the target quantization parameter of each of the encoding units.

12. A video encoding processing device, wherein: include: memory and one or more processors; The memory is used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the video encoding processing method as described in any one of claims 1-10.

13. A non-volatile storage medium storing computer executable instructions, wherein: When the computer executable instructions are executed by a computer processor, they are used to execute the video encoding processing method according to any one of claims 1 to 10.

14. A computer program product comprising a computer program, wherein: When the computer program is executed by a processor, the video encoding processing method according to any one of claims 1 to 10 is implemented.

Images