CN111953966A - Method, device, server and storage medium for testing codes - Google Patents
Method, device, server and storage medium for testing codes Download PDFInfo
- Publication number
- CN111953966A CN111953966A CN202010803701.4A CN202010803701A CN111953966A CN 111953966 A CN111953966 A CN 111953966A CN 202010803701 A CN202010803701 A CN 202010803701A CN 111953966 A CN111953966 A CN 111953966A
- Authority
- CN
- China
- Prior art keywords
- test
- coding
- code rate
- test point
- point set
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000012360 testing method Methods 0.000 title claims abstract description 635
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000013139 quantization Methods 0.000 claims description 89
- 238000004590 computer program Methods 0.000 claims description 5
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 description 12
- 238000010586 diagram Methods 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 5
- 238000013459 approach Methods 0.000 description 3
- 230000003247 decreasing effect Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011056 performance test Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N17/00—Diagnosis, testing or measuring for television systems or their details
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
- H04N19/154—Measured or subjectively estimated visual quality after decoding, e.g. measurement of distortion
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The embodiment of the invention discloses a method, a device, a server and a storage medium for testing codes. Wherein, the method comprises the following steps: determining a first test point set in a reference coding mode according to a code rate test upper limit calibrated by a video sequence to be tested; determining a second test point set in a label alignment coding mode according to the code rate test section where the first test point set is located; and determining a corresponding coding test result according to the coding performance difference after the video sequence to be tested is coded by adopting the first test point set in the reference coding mode and the second test point set in the opposite coding mode respectively. The technical scheme provided by the embodiment of the invention ensures the code rate consistency of the selected test points in different coding modes, and avoids the situation that the selected test points in different coding modes are in different code rate test sections, thereby realizing the accurate test of the coding performance in different coding modes and improving the reliability of the coding test performance.
Description
Technical Field
The embodiment of the invention relates to the field of video processing, in particular to a method, a device, a server and a storage medium for testing coding.
Background
With the rapid development of the internet technology, the video encoding and decoding technology plays an important role in video transmission, and at this time, in order to meet the requirements of different clients on video resolution and image quality, various high-performance video encoders are generally adopted to compress videos so as to reduce the video transmission burden. The currently commonly used Video Coding and decoding standards include h.264, High Efficiency Video Coding (HEVC), VP8, VP9, the first generation Video Coding Standard developed by the open media alliance (AOMedia Video 1, AV1), source Coding Standard (AVs), and AVs2, etc., and at this time, for different Video Coding standards or different Coding configurations under the same Video Coding Standard, there is a difference in the Coding performance of the same Video, so how to test the Video Coding performance under different Video Coding modes (different Video Coding standards or different Coding configurations under the same Video Coding Standard) to select the Video Coding Standard or Coding configuration under the optimal performance is a problem that generally needs to be considered in the Video field.
Currently, when testing the encoding performance of the same video in different video encoding modes, 4 Quantization Parameter (QP) points or Rate points are generally randomly selected in each video encoding mode, and then each video encoding mode is used to encode the video to calculate the video encoding quality and Rate at each QP point or Rate point in each video encoding mode, and further to draw a Rate Distortion (RD) curve in each video encoding mode, so as to calculate BD-rates between different video encoding modes, and analyze the encoding performance in different video encoding modes.
At this time, if QP points are randomly selected in different video coding modes or code Rate points exist in different code Rate intervals, RD curves in different video coding modes also belong to different code Rate intervals, so that performance performances of different video coding modes at the same code Rate cannot be judged, coding performance represented by BD-Rate calculated according to the RD curves in different video coding modes is unreliable, and accuracy of performance testing in different video coding modes is greatly reduced.
Disclosure of Invention
The embodiment of the invention provides a method, a device, a server and a storage medium for testing codes, which ensure the consistency of code rates corresponding to selected test points in different coding modes and improve the reliability and accuracy of the code testing.
In a first aspect, an embodiment of the present invention provides a method for testing coding, where the method includes:
determining a first test point set in a reference coding mode according to a code rate test upper limit calibrated by a video sequence to be tested;
determining a second test point set in a label alignment coding mode according to the code rate test section where the first test point set is located;
and determining a corresponding coding test result according to the coding performance difference of the video sequence to be tested after coding by adopting the first test point set in the reference coding mode and the second test point set in the opposite coding mode.
In a second aspect, an embodiment of the present invention provides an apparatus for testing coding, where the apparatus includes:
the first test point determining module is used for determining a first test point set in a reference coding mode according to a code rate test upper limit calibrated by a video sequence to be tested;
the second test point determining module is used for determining a second test point set in a label alignment coding mode according to the code rate test section where the first test point set is located;
and the coding test module is used for determining a corresponding coding test result according to the coding performance difference of the video sequence to be tested after coding by adopting the first test point set in the reference coding mode and the second test point set in the opposite coding mode.
In a third aspect, an embodiment of the present invention provides a server, where the server includes:
one or more processors;
storage means for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the method for code testing of any embodiment of the present invention.
In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to implement the method for testing coding according to any embodiment of the present invention.
In the method, the device, the server and the storage medium for testing the coding, provided by the embodiment of the invention, for any coding mode, with the continuous increase of the video coding rate, the video image quality is finally close to lossless, so that the coding test of different coding modes under the over-high coding rate cannot accurately represent the real coding performance of the coding mode, the corresponding upper limit of the coding rate test can be calibrated for the video sequence to be tested in advance, the first test point set under the reference coding mode is determined, the coding rate corresponding to each test point in the first test point set is lower than the upper limit of the coding rate test, the second test point set under the opposite coding mode is determined according to the coding rate test segment where the first test point set is located, and the coding rates corresponding to the test points in the first test point set and the second test point set are located in the same coding rate test segment, the code rate corresponding to each test point in the second test point set is also lower than the upper limit of the code rate test, the consistency of the code rates corresponding to the selected test points in different coding modes is ensured, and the situation that the selected test points in different coding modes are in different code rate test sections is avoided, so that the accurate test of the coding performance in different coding modes is realized, and the reliability of the coding test performance is improved.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments made with reference to the following drawings:
fig. 1A is a flowchart of a method for testing codes according to an embodiment of the present invention;
fig. 1B is a schematic diagram of a coding test process according to an embodiment of the present invention;
fig. 2A is a flowchart of a method for testing encoding according to a second embodiment of the present invention;
fig. 2B is a schematic diagram of a coding test process according to a second embodiment of the present invention;
fig. 3A is a flowchart of a method for testing encoding according to a third embodiment of the present invention;
fig. 3B is a schematic diagram of a coding test process according to a third embodiment of the present invention;
fig. 3C is a schematic diagram of RD curves in the reference coding mode and the benchmarking coding mode in the method according to the third embodiment of the present invention.
Fig. 4 is a schematic structural diagram of a transcoding device for live video according to a fourth embodiment of the present invention;
fig. 5 is a schematic structural diagram of a server according to a fifth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures. In addition, the embodiments and features of the embodiments in the present invention may be combined with each other without conflict.
Example one
Fig. 1A is a flowchart of a method for coding test according to an embodiment of the present invention, which can be applied to a case where an encoding mode with optimal coding performance is selected for a certain video to be encoded. The method for testing the coding provided by this embodiment may be performed by an apparatus for testing the coding provided by the embodiment of the present invention, which may be implemented in a software and/or hardware manner, and is integrated in a server for performing the method, where the server may be a background server for various video applications.
Specifically, referring to fig. 1A, the method may include the steps of:
s110, determining a first test point set in a reference coding mode according to the code rate test upper limit calibrated by the video sequence to be tested.
Specifically, because different video coding standards (i.e., encoders) or different coding configurations under the same video coding standard are adopted, corresponding performance differences exist when the same video is coded, at this time, in this embodiment, coding configurations under each video coding standard or video coding standard that may be adopted by video coding are taken as corresponding coding modes, so that coding tests are conveniently performed on the same video under different coding modes, and the video sequence to be tested can be a complete video that needs to be currently subjected to the coding test to select the coding mode under the optimal coding performance to perform coding, or a video segment that is randomly extracted from the complete video.
Furthermore, for any coding mode, along with the continuous increase of the video coding rate, the improvement of the video coding quality caused by the continuous increase of the video coding rate gradually becomes slow, and finally the video quality can reach or approach lossless video image quality, namely the improvement of the video coding quality under any coding mode gradually tends to be smooth along with the increase of the coding rate, so that the video coding quality after being coded under the overhigh coding rate of different coding modes approaches the lossless video image quality, belongs to the inherent characteristic of the video coding under the overhigh coding rate and cannot accurately represent the real coding performance of each coding mode, therefore, through analyzing the historical coding experience, the corresponding upper limit of the code rate test can be calibrated for the video sequence to be tested in advance, and the code rate corresponding to each test point selected when the subsequent coding test is carried out under different coding modes is required to be lower than the upper limit of the code rate test, the method ensures that the video coding quality obtained after the coding test is carried out in different coding modes can accurately represent the real coding performance of the coding mode, thereby improving the test reliability in different coding modes.
It should be noted that the test point selected in this embodiment may be a code Rate point, or may also be a QP point or a Constant Code Rate Factor (CRF) point that has a certain relationship with a code Rate, and the specific need is determined according to a coding configuration condition.
Meanwhile, in order to ensure the accuracy of the coding test, when the corresponding test points are used to analyze the coding performance of the video sequence to be tested in different coding modes, the fairness of the test points used between the different coding modes needs to be ensured first, therefore, the test point adopted in a specific coding mode of different coding modes can be used as a reference to select the test points in other coding modes, the referenced coding mode is the reference coding mode in this embodiment, and the other coding modes to be tested are the opposite sign coding modes in this embodiment, for example, when testing the difference of coding performance when coding the video sequence to be tested by respectively adopting the coding configuration cfgA under the coder A and the coding configuration cfgB under the coder B, the encoder a and the coding configuration cfgA therebelow can be taken as a reference coding mode, and the encoder B and the coding configuration cfgB therebelow can be taken as a parity coding mode.
In this embodiment, when the coding performance of a video sequence to be tested in different coding modes needs to be analyzed, a corresponding code rate test upper limit is calibrated for the video sequence to be tested in advance through historical coding experiences at different coding code rates, and when the video sequence to be tested is required to be tested in different coding modes, the code rate corresponding to the selected test point in each coding mode is not allowed to exceed the code rate test upper limit, so as to ensure the reliability of the coding performance test in different coding modes; furthermore, firstly, referring to the code rate test upper limit calibrated by the video sequence to be tested, screening out the test points under the target quantity according to a certain rule from the corresponding test points in the code rate range less than or equal to the code rate test upper limit, so as to combine and obtain the first test point set under the reference coding mode, wherein the code rate corresponding to each test point in the first test point set is less than or equal to the code rate test upper limit, that is, the video coding quality after the video sequence to be tested is coded by adopting the first test point set under the reference coding mode can accurately represent the real coding performance of the reference coding mode.
And S120, determining a second test point set in the label alignment coding mode according to the code rate test section where the first test point set is located.
Optionally, when the same video is coded by different code rate segments in any coding mode, the coding performance of the different code rate segments also has a certain difference, so that the coding performance of the video in other code rate segments cannot be predicted by the coding performance of a certain code rate segment in the coding mode, therefore, when the coding performance of the video sequence to be tested in different coding modes is tested, in order to ensure the fairness of the coding performance test in different coding modes, the code rates corresponding to the selected test points in the reference coding mode and the target coding mode are required to be in the same code rate segment, so that the test conditions in different coding modes are consistent.
It can be understood that, after the first test point set in the reference coding mode is determined, in order to make the code rates corresponding to the test points selected in the reference coding mode and the benchmarking coding mode to be in the same code rate segment, first, a relationship between the test points in the first test point set and code rate values needs to be analyzed to calculate the code rate value corresponding to each test point in the first test point set, as shown in fig. 1B, so as to determine the code rate test segment where the first test point set is located; then, the video sequence to be tested is coded for a plurality of times by continuously adopting a mark-aligning coding mode with reference to a code rate test section where the first test point set is located, the adopted test point values are continuously updated in the coding process for a plurality of times so that the coding code rate in the mark-aligning coding mode is continuously close to the code rate test section where the first test point set is located, thereby obtaining a second test point set in the mark-aligning coding mode, the code rates corresponding to the second test point set and each test point in the first test point set are located in the same code rate test section, the code rate corresponding to each test point in the second test point set is also ensured to be less than or equal to the upper limit of the code rate test marked by the video sequence to be tested, the consistency of the code rates corresponding to the test points selected under the benchmark coding mode and the mark-aligning coding mode is ensured, and the performance difference after the video sequence to be tested is coded in the same code rate test section by adopting the benchmark coding mode and, the situation that the selected test points are in the test segments with different code rates in different coding modes is avoided, so that the accurate test of the coding performance in different coding modes is realized, and the reliability of the coding test performance is improved.
For example, in this embodiment, the code rate test segment where the first test point set is located may be directly used as the code rate test segment in the mark aligning coding mode, and the second test point set in the mark aligning coding mode may be calculated according to the start-stop code rate in the code rate test segment. That is, the initial code rate and the termination code rate in the code rate test section are determined, then the test points corresponding to the initial code rate and the termination code rate are respectively calculated according to the relationship between specific test points and code rate values, and then the corresponding other test points are searched between the two test points corresponding to the initial code rate and the termination code rate according to the test point distribution rule in the first test point set, so that the second test point set in the mark-matching coding mode is obtained through combination. For example, if there are 4 test points at fixed intervals in the first test point set, then 2 test points are also searched between two test points corresponding to the start code rate and the stop code rate according to the corresponding intervals, and the two test points corresponding to the start code rate and the stop code rate jointly form a second test point set including 4 test points in the mark-pair coding mode.
S130, according to the coding performance difference after the video sequence to be tested respectively adopts the first test point set in the reference coding mode and the second test point set in the opposite coding mode, determining the corresponding coding test result.
Optionally, after determining a first test point set in a reference coding mode and a second test point set in a label matching coding mode, the reference coding mode may be utilized, and each test point in the first test point set is used to code the video sequence to be detected, and the video coding quality of the video sequence to be detected at each test point in the first test point set is calculated, so as to analyze the change condition of the video coding quality along with the increase of the code rate corresponding to the first test point set, so as to obtain the video coding performance of the video sequence to be detected in the code rate test segment where the first test point set is located when the video sequence to be detected is coded in the reference coding mode; meanwhile, a mark-aligning coding mode is utilized, each test point in the second test point set is adopted to code the video sequence to be tested, the video coding quality of the video sequence to be tested under each test point in the second test point set is calculated, the change condition of the video coding quality along with the increase of the corresponding code rate of the second test point set is further analyzed, the video coding performance of the video sequence to be tested in a code rate test section in which the second test point set is positioned when the mark-aligning coding mode is adopted to code the video sequence to be tested is obtained, at the moment, the code rate test section in which the first test point set is positioned is consistent with the code rate test section in which the second test point set is positioned, therefore, the corresponding coding performance difference is determined by comparing the coding performance of the video sequence to be tested under the reference coding mode and the mark-aligning coding mode in the code rate test section, and the coding mode with the optimal coding performance is further selected from the reference coding mode and the mark-, and the code test result corresponding to the current code test is obtained, so that the reliability of the code test result is improved.
In the technical scheme provided by this embodiment, as the video coding rate of any coding mode is continuously increased, the video quality is finally close to lossless video quality, so that the coding test of different coding modes under an excessively high code rate cannot accurately represent the real coding performance of the video sequence, a corresponding code rate test upper limit can be calibrated for a video sequence to be tested in advance, a first test point set under a reference coding mode is determined, the code rate corresponding to each test point in the first test point set is lower than the code rate test upper limit, a second test point set under the benchmarking coding mode is determined according to the code rate test section where the first test point set is located, the code rates corresponding to the test points in the first test point set and the second test point set are located in the same code rate test section, that is, the code rate corresponding to each test point in the second test point set is also lower than the code rate test upper limit, and the consistency of the code rates corresponding to the selected test points in different coding modes is ensured, and the condition that the selected test points in different coding modes are in different code rate test sections is avoided, so that the accurate test of the coding performance in different coding modes is realized, and the reliability of the coding test performance is improved.
Example two
Fig. 2A is a flowchart of a method for testing codes according to a second embodiment of the present invention, and fig. 2B is a schematic diagram of a principle of a process for testing codes according to the second embodiment of the present invention. The embodiment is optimized on the basis of the embodiment. Specifically, the present embodiment mainly explains a specific determination process of the first test point set in the reference coding mode and the second test point set in the target coding mode in detail.
Optionally, as shown in fig. 2A, this embodiment may include the following steps:
s201, determining the types of the test points in the first test point set and the second test point set according to the code control configuration conditions of the reference coding mode and the opposite coding mode.
Optionally, because different configuration requirements may exist in the corresponding coding mode according to different coding requirements, for example, when there is a specific requirement for a coding Rate, corresponding code control configuration conditions, such as Average code Rate (ABR), dynamic code Rate (VBR), Constant code Rate (CBR), and the like, may be preset in the coding mode, and when there is a specific requirement for coding quality, there may be no corresponding code control configuration conditions, and quality conditions, such as Constant Quantization Parameter (CQP) and Constant Code Rate Factor (CRF), may be set, and the like, so to improve the adaptability between the test point and the coding mode, in this embodiment, first, the configuration parameters in the reference coding mode and the target coding mode are analyzed, whether there is a corresponding code control configuration condition is determined, and then, the types of the test points in the first test point set and the second test point set are determined, so that the test points under the corresponding types can be directly selected in the following.
It should be noted that, in order to ensure fairness of the coding test, the embodiment also requires that code control configuration conditions of the reference coding mode and the target coding mode are consistent.
For example, if the code control configuration condition exists in the reference coding mode and the label matching coding mode, the types of the test points in the first test point set and the second test point set are determined to be code rate points, and if the code control configuration condition does not exist in the reference coding mode and the label matching coding mode, the types of the test points in the first test point set and the second test point set are determined to be QP points, where the CRF point is a special QP point.
S202, judging whether the type of the test point is a code rate point or not; if yes, go to S203; if not, go to step S204.
Optionally, because the types of the test points are different, the specific steps executed when determining the first test point set in the reference coding mode and the second test point set in the opposite coding mode are also different, and therefore it is necessary to determine whether the type of the test point is a code rate point to execute different test point determining steps.
S203, determining a first test point set in a reference coding mode according to the code rate test upper limit calibrated by the video sequence to be tested, wherein the second test point set is the same as the first test point set.
Optionally, if the type of the test point is a code rate point, a plurality of code rate points corresponding to specific code rate values less than or equal to the code rate test upper limit may be directly determined according to the code rate test upper limit calibrated by the video sequence to be tested, and the code rate points are used as the test points in the first test point set in the reference coding mode, at this time, the first test point set may be used as the second test point set in the mark-aligning coding mode, so that the reference coding mode and the mark-aligning coding mode perform coding test on the same code rate point, thereby ensuring the accuracy of the coding test.
For example, in this embodiment, a code rate test upper limit (defined as bitrate _ anchor) calibrated by the video sequence to be tested may be directly used as a maximum code rate point in the first test point set in the reference coding mode, and then the maximum code rate point is sequentially reduced according to a fixed interval to obtain other code rate points in the first test point set; if there are 4 test points in the first test point set, the 4 test points may be bitrate _ anchor/4, bitrate _ anchor/2, bitrate _ anchor 3/4 and bitrate _ anchor, respectively, and the test points in the second test point set are bitrate _ anchor/4, bitrate _ anchor/2, bitrate _ anchor 3/4 and bitrate _ anchor.
And S204, coding the video sequence to be detected by adopting the default quantization parameter in the reference coding mode.
Optionally, if the type of the test point is a non-rate point, that is, the test point in the first test point set and the second test point set is required to be a QP point, when the first test point set in the reference coding mode is determined, first, a default quantization parameter (for example, QP ═ 22) in the reference coding mode may be used to encode the video sequence to be tested, so as to determine whether the coded rate corresponding to the default quantization parameter meets the requirement of the upper limit of the rate test specified by the video sequence to be tested, that is, determine whether the coded rate is less than or equal to the upper limit of the rate test, at this time, since the QP value may represent the detail compression condition of the video picture in the coding process, if the QP is smaller, most of the detail in the video picture is retained, and if the QP is increased, part of the detail is lost, that is, the QP value and the rate are in an inverse ratio relationship, the code rate after coding corresponding to the default quantization parameter can be updated correspondingly by continuously updating the default quantization parameter, and then the quantization parameter corresponding to the code rate after coding allowed to be reached in the reference coding mode is determined by judging the size between the code rate after coding and the code rate test upper limit calibrated by the video sequence to be tested, so that the quantization point in the first test point set in the reference coding mode is determined.
S205, if the code rate after coding is larger than the code rate test upper limit marked by the video sequence to be tested, updating the default quantization parameter, continuing to code the video sequence to be tested until the code rate after coding is smaller than or equal to the code rate test upper limit, and taking the latest default quantization parameter as the minimum test point in the first test point set.
Optionally, after the video sequence to be detected is encoded by using the default quantization parameter in the reference encoding mode, if the obtained code rate after coding is larger than the code rate test upper limit marked by the video sequence to be tested, the code rate is not in accordance with the requirement of the code rate test upper limit, therefore, the default quantization parameter can be updated according to the code rate difference between the currently obtained code rate after coding and the code rate test upper limit, since the QP value is inversely proportional to the code rate, the default quantization parameter is updated accordingly, then adopting the increased default quantization parameter in the reference coding mode to continuously code the video sequence to be tested, circulating in sequence, and continuously reducing the code rate after coding until the code rate is less than or equal to the upper limit of the code rate test, and taking the latest default quantization parameter as the minimum test point in the first test point set.
S206, determining other test points in the first test point set based on the minimum test point and a preset quantization point rule.
Optionally, after the minimum test point in the first test point set is obtained, in order to make the code rates corresponding to the test points in the first test point set all smaller than the code rate test upper limit, the minimum test point is continuously increased, so that the corresponding code rates are continuously decreased, at this time, the preset quantization point rule may be a fixed quantization interval that is increased each time, that is, the fixed quantization interval is sequentially increased at the minimum test point, and new quantization points are continuously obtained and used as other test points in the first test point set.
Illustratively, if the smallest test point in the first set of test points is a QPA1The fixed quantization interval is 5, 4 test points are in the first test point set, and then other test points in the first test point set are respectively QPA2＝QPA1+5、QPA3＝QPA1+10 and QPA4＝QPA1+15。
At this time, the second test point set in the mark-to-mark coding mode may be calculated according to the code rate test section in which the first test point set is located, and the initial code rate in the code rate test section in which the first test point set is located may be a code rate after coding the video sequence to be tested by using the largest test point in the first test point set, and the end code rate may be a code rate after coding the video sequence to be tested by using the smallest test point in the first test point set.
And S207, coding the video sequence to be detected by adopting a default initial quantization point in a mark coding mode.
Optionally, since the present embodiment requires that the code rate test segments where the first test point set and the second test point set are located are the same, the specific quantization value of the maximum test point corresponding to the start code rate in the second test point set and the specific quantization value of the minimum test point corresponding to the end code rate may be analyzed according to the start-stop code rate in the code rate test segment where the first test point set is located.
At this time, when analyzing the minimum test point in the second test point set, based on the requirement of consistency of the code rate test segments where the first test point set and the second test point set are located, the minimum test point in the first test point set may be first used as a default initial quantization point in the target coding mode in this embodiment, then the default initial quantization point in the target coding mode is used to code the video sequence to be tested, and it is determined whether the obtained coded code rate exceeds the close range of the termination code rate in the code rate test segment where the first test point set is located, where the close range can represent the error range that ensures the allowed termination code rate when the code rate test segments where the first test point set and the second test point set are located are consistent, so as to adjust the coded code rate by updating the default initial quantization point, and as far as possible, the code rate corresponding to the minimum test point in the second test point set gradually approaches the final code rate test segment where the first test point set is located And (5) stopping code rate.
And S208, if the code rate after coding exceeds the adjacent range of the termination code rate in the code rate test section where the first test point set is located, updating the default initial quantization point, continuing coding the video sequence to be tested until the code rate after coding is in the adjacent range of the termination code rate, and taking the latest default initial quantization point as the minimum test point in the second test point set.
Optionally, after the video sequence to be tested is encoded by using the default initial quantization point in the mark-pair encoding mode, if the encoded code rate exceeds the proximity range of the termination code rate in the code rate test segment where the first test point set is located, it is described that the encoded code rate is not consistent with the termination code rate in the code rate test segment where the first test point set is located within the error tolerance range, so that the default initial quantization point needs to be updated according to the size between the encoded code rate and the termination code rate in the code rate test segment where the first test point set is located, for example, if the encoded code rate is greater than the termination code rate in the code rate test segment where the first test point set is located, the default initial quantization point is increased, and if the encoded code rate is less than or equal to the termination code rate in the code rate test segment where the first test point set is located, the default initial quantization point is decreased, and then the updated default initial quantization point in the mark-pair encoding mode is used to continue to encode the video sequence to be tested And coding and sequentially circulating until the newly obtained coded code rate is in the close range of the termination code rate, which indicates that the coded code rate corresponding to the latest default initial quantization point is close to the termination code rate in the code rate test section in which the first test point set is positioned, and is consistent with the termination code rate within the error allowable range, so that the latest default initial quantization point is used as the minimum test point in the second test point set, and the accuracy of the minimum test point in the second test point set is ensured.
For example, in this embodiment, in order to accurately determine whether the coded bit rate is in the close range of the termination bit rate in the bit rate test segment where the first test point set is located, two parameters, i.e., flag1 and flag2, are preset, and the flag1 is initially defined as when flag2 is 0, at this time, the values of flag1 and flag2 are changed by determining the size between the coded bit rate and the termination bit rate in the bit rate test segment where the first test point set is located, when coding is started, the flag1 is defined as when flag2 is 0, if the coded bit rate is greater than the termination bit rate in the bit rate test segment where the first test point set is located, the flag1 is set to 1, then the default starting quantization point is continuously updated, when the video sequence to be coded is continuously reduced, until the first coded bit rate is less than or equal to the termination bit rate in the bit rate test segment where the first test point set is located after coding, the flag2 is set to 1, at this time, the updating of the default initial quantization point is explained to enable the code rate after coding to firstly cross the termination code rate in the code rate test section where the first test point set is located; or when the encoding is just started, setting flag1 to flag2 to 0, if the code rate after encoding is less than or equal to the termination code rate in the code rate test section where the first test point set is located, setting flag2 to 1, then continuously increasing the code rate after encoding when the video sequence to be detected is encoded continuously by continuously updating the default starting quantization point until the code rate after encoding is first greater than the termination code rate in the code rate test section where the first test point set is located, setting flag1 to 1, and at this time, explaining that the updating of the default starting quantization point makes the code rate after encoding first cross the termination code rate in the code rate test section where the first test point set is located; that is, only when the flag1+ flag2 is 2, it indicates that the current post-coding code rate is in the close range of the termination code rate in the code rate test segment where the first test point set is located, and when the flag1+ flag2 is <2, it indicates that the current post-coding code rate exceeds the close range of the termination code rate in the code rate test segment where the first test point set is located, and the default starting quantization point needs to be updated continuously.
Accordingly, in order to reduce the number of times of repeated encoding of the video sequence to be detected, before updating the default starting quantization point, the embodiment may further include: and determining the current updating coefficient of the default initial quantization point according to the code rate difference between the code rate after coding and the code rate after termination.
Specifically, different update coefficients are set according to the code rate difference between the code rate after coding and the code rate after termination to correspondingly update the default initial quantization point, so that the update amplitude of the default initial quantization point is different under different code rate differences, the current update coefficient of the default initial quantization point is determined according to the code rate difference between the code rate after coding and the code rate after termination when the default initial quantization point needs to be updated each time, and the current update coefficient is set to be larger when the code rate difference is larger, so that repeated coding under the unnecessary default initial quantization point is reduced; for example, if the code rate difference between the encoded code rate and the end code rate exceeds 6%, the current update coefficient QPoffs _ lower of the default starting quantization point may be set to 3, and if the code rate difference is between 3% -6%, the current update coefficient QPoffs _ lower of the default starting quantization point may be set to 2, and so on.
S209, the video sequence to be detected is coded by adopting the default termination quantization point in the encoding mode.
Optionally, when analyzing the largest test point in the second test point set, the same logic as the analysis of the smallest test point in the second test point set is used, first, the largest test point in the first test point set is used as the default termination quantization point in the target coding mode in this embodiment, then, the default termination quantization point in the mark coding mode is adopted to code the video sequence to be tested, whether the obtained code rate after coding exceeds the adjacent range of the initial code rate in the code rate test segment in which the first test point set is positioned is judged, the proximity range can represent an error range of the allowed initial code rate when the code rate test segments of the first test point set and the second test point set are consistent, so as to adjust the code rate after coding by updating the default termination quantization point, and gradually make the code rate corresponding to the maximum test point in the second test point set close to the initial code rate in the code rate test section where the first test point set is located as far as possible.
S210, if the code rate after coding exceeds the adjacent range of the initial code rate in the code rate test section where the first test point set is located, updating the default termination quantization point, continuing coding the video sequence to be tested until the code rate after coding is in the adjacent range of the initial code rate, and taking the latest default termination quantization point as the maximum test point in the second test point set.
Optionally, after the video sequence to be tested is encoded by using the default termination quantization point in the mark-pair encoding mode, if the encoded code rate exceeds the proximity range of the initial code rate in the code rate test segment where the first test point set is located, it is described that the encoded code rate is inconsistent with the initial code rate in the code rate test segment where the first test point set is located within the error tolerance range, so that the default termination quantization point needs to be updated according to the size between the encoded code rate and the initial code rate in the code rate test segment where the first test point set is located, for example, if the encoded code rate is greater than the initial code rate in the code rate test segment where the first test point set is located, the default termination quantization point is increased, and if the encoded code rate is less than or equal to the initial code rate in the code rate test segment where the first test point set is located, the default termination quantization point is decreased, and then the updated default termination quantization point in the mark-pair encoding mode is used to continue to encode the video sequence to be tested And coding and sequentially circulating until the newly obtained coded code rate is in the range close to the initial code rate, which shows that the coded code rate corresponding to the latest default termination quantization point is close to the initial code rate in the code rate test section in which the first test point set is positioned and is consistent with the initial code rate in the error allowable range, so that the latest default termination quantization point is used as the maximum test point in the second test point set, and the accuracy of the maximum test point in the second test point set is ensured.
Correspondingly, in order to reduce the number of repeated coding of the video sequence to be detected, before the default termination quantization point is updated, the operation is the same as the updating operation of the default starting quantization point, and the current updating coefficient of the default termination quantization point is also determined according to the code rate difference between the coded code rate and the starting code rate.
It should be noted that, in this embodiment, S207 to S208 are processes of calculating a minimum test point in the second test point set in the standard coding mode according to the first test point set in the reference coding mode, S209 to S210 are processes of calculating a maximum test point in the second test point set in the standard coding mode according to the first test point set in the reference coding mode, and no fixed execution order exists between S207 to S208 and S209 to S210, so that the minimum test point and the maximum test point in the second test point set in the standard coding mode can be calculated at the same time.
S211, determining other test points in the second test point set based on the minimum test point and the maximum test point in the second test point set.
Optionally, after obtaining the minimum test point and the maximum test point in the second test point set, this embodiment refers to a preset test point distribution rule, and selects another test point in the second test point set from quantization points between the minimum test point and the maximum test point.
For example, if there are 4 test points in the second set of test points, the minimum test point is QPB1The maximum test point is QPB4Then the second set of test points do not divide QPB1And QPB4In addition, QP is includedB1And QPB4QP betweenB2And QPB3At this time QPB2＝QPB1*2/3+QPB4/3，QPB3＝QPB1/3+QPB4*2/3。
S212, according to the coding performance difference after the video sequence to be tested respectively adopts the first test point set in the reference coding mode and the second test point set in the opposite coding mode, the corresponding coding test result is determined.
The technical solution provided in this embodiment determines the types of the test points in the first test point set and the second test point set according to the code control configuration conditions of the reference coding mode and the peer-to-peer coding mode, so as to achieve the adaptability between the test points in the first test point set and the second test point set and the code test, and at the same time, repeatedly codes the video sequence to be tested by using the reference coding mode or the peer-to-peer coding mode by continuously updating the default quantization point used for coding the video sequence to be tested, so as to analyze the consistency of the corresponding code rates of the test points in the first test point set and the second test point set, so that the corresponding code rate of each test point in the first test point set and the second test point set is lower than the upper limit of the code rate test, ensure the consistency of the corresponding code rates of the selected test points in different coding modes, and avoid the situation that the selected test points in different coding modes are in, therefore, the coding performance can be accurately tested in different coding modes, and the reliability of the coding test performance is improved.
EXAMPLE III
Fig. 3A is a flowchart of a method for testing codes according to a third embodiment of the present invention, and fig. 3B is a schematic diagram of a principle of a process for testing codes according to the third embodiment of the present invention. The embodiment is optimized on the basis of the embodiment. Specifically, the present embodiment mainly explains the debugging process of the video sequence to be tested and the specific analysis process of the coding performance difference of the video sequence to be tested in the reference coding mode and the opposite coding mode in detail.
Optionally, as shown in fig. 3A, this embodiment may include the following steps:
s310, a reference coding mode and a benchmarking coding mode are screened out from a coding mode library to be tested.
Optionally, the encoding mode library to be tested may include various existing video encoding standards and various encoding configurations thereof under the video encoding standards, and each video encoding standard or each encoding configuration under the video encoding standard may be used as an encoding mode, so when an encoding mode with optimal encoding performance is selected for a video sequence to be tested, a reference encoding mode and a target encoding mode that need to be subjected to encoding test facing the video sequence to be tested need to be screened from the encoding mode library to be tested.
S320, if the video sequence to be tested is in a non-compressed format, encoding the video sequence to be tested by adopting default quantization parameters in a reference encoding mode, and taking the code rate after encoding as the code rate test upper limit calibrated by the video sequence to be tested, otherwise, taking the original code rate after the video sequence to be tested is compressed as the code rate test upper limit calibrated by the video sequence to be tested.
Optionally, when the upper limit of the code rate test calibrated by the video sequence to be tested is judged, because the video sequence to be tested oriented to the coding in this embodiment is usually a video reported to the background server by a certain client, some original coding information can be carried, and at this time, the original code rate of the video sequence to be tested after being compressed can be directly used as the upper limit of the code rate test calibrated by the video sequence to be tested with reference to the original coding condition of the video sequence to be tested; however, if the video sequence to be tested is in a non-compressed format, that is, the video sequence to be tested is not originally encoded before, the embodiment may first perform preliminary encoding on the video sequence to be tested by using the default quantization parameter in the reference encoding mode, where the default quantization parameter may be adjusted according to an actual encoding situation, and then the encoded code rate is used as a code rate test upper limit calibrated for the video sequence to be tested, thereby ensuring the adaptivity of the code rate test upper limit.
For example, if the video sequence to be tested is stored in the color-coded YUV format, then with the reference coding mode, the QP is selected to be 22 in the CQP configuration for one-time coding.
S330, if the video sequence to be detected is in a compressed format, converting the video sequence to be detected into a non-compressed format, and setting the size of the video sequence to be detected according to the lower limit of the coding size.
Optionally, when the video sequence to be tested is subjected to encoding test, the video sequence to be tested is required to be a set of decapsulated video frames, so if the video sequence to be tested is in a compressed format, the video sequence to be tested needs to be converted into an uncompressed format, for example, the video sequence to be tested is converted into a YUV format by using an ffmpeg technology.
Meanwhile, considering that the minimum size of the coding unit used when the video sequence to be detected is coded is 8 × 8, in order to avoid coding errors of the video sequence to be detected when the video sequence to be detected does not conform to the size, in this embodiment, the lower limit of the coding size is set to 8 × 8, and the size of the video sequence to be detected needs to be set to a multiple of the lower limit of the coding size, for example, the video sequence to be detected is converted into a YUV format with a length and a width both being multiples of 8 by using ffmpeg technology, at this time, if the original length of the video sequence to be detected is width _ org and the original width is height _ org, the length of the YUV image converted by the ffmpeg technology may be (int) (width _ org/8) × 8, and the width may be (int) (height _ org/8).
S340, determining a first test point set in the reference coding mode according to the code rate test upper limit calibrated by the video sequence to be tested.
And S350, determining a second test point set in the label alignment coding mode according to the code rate test section where the first test point set is located.
And S360, coding the video sequence to be tested by respectively adopting the first test point set in the reference coding mode and the second test point set in the opposite coding mode.
Optionally, after determining the first test point set in the reference coding mode and the second test point set in the label coding mode, for the coding test of the video sequence to be tested in the reference coding mode and the label coding mode, firstly, each test point in the first test point set in the reference coding mode is used for coding the video sequence to be tested, meanwhile, each test point in the second test point set in the label coding mode is used for coding the video sequence to be tested, and then, the video coding quality of each test point in the reference coding mode and the label coding mode is analyzed, so that an accurate coding test result is obtained.
And S370, respectively generating Rate-Distortion (RD) curves of the video sequence to be detected in the reference coding mode and the opposite coding mode by adopting a preset third-party coding evaluation tool, and calculating corresponding coding performance deviation.
Optionally, after the video sequence to be tested is coded at each test point in the reference coding mode and the target coding mode, a code stream at each test point is obtained, and then after each code stream is decoded, a preset third-party coding evaluation tool is used to compare the decoding result of each code stream with the original video sequence to be tested, so as to calculate the video coding quality corresponding to each test point, such as Peak Signal-to-Noise Ratio (PSNR) or Structural SIMilarity (SSIM), and the like, thereby avoiding that the fairness of the coding test results is influenced by the difference in details of methods used in calculating the guest quality data in different coding modes; furthermore, the video coding quality of each test point in the reference coding mode and the target coding mode is referred to, the video coding quality of each test point in the first test point set and the video coding quality of each test point in the second test point set are fitted in the reference coding mode and the target coding mode, so as to generate RD curves of the video sequence to be tested in the reference coding mode and the target coding mode, as shown in fig. 3C, and then the corresponding BD-rate is calculated according to the coding performance difference between the RD curves in the reference coding mode and the target coding mode, as the coding performance deviation in this embodiment, so that the corresponding coding test result is determined based on the coding performance deviation in the following, thereby ensuring the accuracy of the coding test.
And S380, determining a corresponding coding test result based on the coding performance deviation.
Optionally, based on the calculated coding performance deviation (BD-rate), the degree of optimization of the coding performance of the video sequence to be tested in the reference coding mode and the opposite coding mode is determined, so that the coding mode with the optimal coding performance is screened out as the corresponding coding test result.
The technical solution provided in this embodiment is to determine the compression format and size of a video sequence to be tested, to ensure the adaptivity of the coding test of the video sequence to be tested, then to calibrate the corresponding upper limit of the code rate test for the video sequence to be tested in advance, and to determine the first test point set in the reference coding mode, so that the code rate corresponding to each test point in the first test point set is lower than the upper limit of the code rate test, and to determine the second test point set in the reference coding mode according to the code rate test segment where the first test point set is located, so that the code rates corresponding to the test points in the first test point set and the second test point set are located in the same code rate test segment, that is, the code rate corresponding to each test point in the second test point set is also lower than the upper limit of the code rate test, and to ensure the consistency of the code rates, the situation that the selected test points are in the test segments with different code rates in different coding modes is avoided, so that the accurate test of the coding performance in different coding modes is realized, and the reliability of the coding test performance is improved.
Example four
Fig. 4 is a schematic structural diagram of an apparatus for testing encoding according to a fourth embodiment of the present invention, specifically, as shown in fig. 4, the apparatus may include:
a first test point determining module 410, configured to determine a first test point set in a reference coding mode according to a code rate test upper limit calibrated by a video sequence to be tested;
a second test point determining module 420, configured to determine a second test point set in the label alignment coding mode according to the code rate test segment where the first test point set is located;
the encoding test module 430 is configured to determine a corresponding encoding test result according to the encoding performance difference of the video sequence to be tested after encoding the video sequence respectively using the first test point set in the reference encoding mode and the second test point set in the opposite encoding mode.
In the technical scheme provided by this embodiment, as the video coding rate of any coding mode is continuously increased, the video quality is finally close to lossless video quality, so that the coding test of different coding modes under an excessively high code rate cannot accurately represent the real coding performance of the video sequence, a corresponding code rate test upper limit can be calibrated for a video sequence to be tested in advance, a first test point set under a reference coding mode is determined, the code rate corresponding to each test point in the first test point set is lower than the code rate test upper limit, a second test point set under the benchmarking coding mode is determined according to the code rate test section where the first test point set is located, the code rates corresponding to the test points in the first test point set and the second test point set are located in the same code rate test section, that is, the code rate corresponding to each test point in the second test point set is also lower than the code rate test upper limit, and the consistency of the code rates corresponding to the selected test points in different coding modes is ensured, and the condition that the selected test points in different coding modes are in different code rate test sections is avoided, so that the accurate test of the coding performance in different coding modes is realized, and the reliability of the coding test performance is improved.
The device for testing the codes provided by the embodiment can be suitable for the method for testing the codes provided by any embodiment, and has corresponding functions and beneficial effects.
EXAMPLE five
Fig. 5 is a schematic structural diagram of a server according to a fifth embodiment of the present invention, and as shown in fig. 5, the server includes a processor 50, a storage device 51, and a communication device 52; the number of the processors 50 in the server may be one or more, and one processor 50 is taken as an example in fig. 5; the processor 50, the storage device 51 and the communication device 52 in the server may be connected by a bus or other means, and the bus connection is taken as an example in fig. 5.
The server provided by the embodiment can be used for executing the method for testing the codes provided by any embodiment, and has corresponding functions and beneficial effects.
EXAMPLE six
An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, can implement the method for testing coding in any of the above embodiments. The method specifically comprises the following steps:
determining a first test point set in a reference coding mode according to a code rate test upper limit calibrated by a video sequence to be tested;
determining a second test point set in a label alignment coding mode according to the code rate test section where the first test point set is located;
and determining a corresponding coding test result according to the coding performance difference of the video sequence to be tested after coding by adopting the first test point set in the reference coding mode and the second test point set in the opposite coding mode.
Of course, the storage medium provided by the embodiment of the present invention contains computer-executable instructions, and the computer-executable instructions are not limited to the method operations described above, and may also perform related operations in the method for code testing provided by any embodiment of the present invention.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the methods according to the embodiments of the present invention.
It should be noted that, in the embodiment of the apparatus for testing coding, the included units and modules are merely divided according to functional logic, but are not limited to the above division as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (15)
1. A method of code testing, comprising:
determining a first test point set in a reference coding mode according to a code rate test upper limit calibrated by a video sequence to be tested;
determining a second test point set in a label alignment coding mode according to the code rate test section where the first test point set is located;
and determining a corresponding coding test result according to the coding performance difference of the video sequence to be tested after coding by adopting the first test point set in the reference coding mode and the second test point set in the opposite coding mode.
2. The method of claim 1, wherein the determining the second set of test points in the parity coding mode according to the code rate test segment in which the first set of test points is located comprises:
and taking the code rate test section where the first test point set is located as the code rate test section in the label alignment coding mode, and calculating a second test point set in the label alignment coding mode according to the start-stop code rate in the code rate test section.
3. The method of claim 1, prior to determining the first set of test points in the reference coding mode, further comprising:
and determining the types of the test points in the first test point set and the second test point set according to the code control configuration conditions of the reference coding mode and the label alignment coding mode.
4. The method of claim 3, wherein if the test points in the first test point set are quantization points, the determining the first test point set in the reference coding mode according to the upper limit of the code rate test calibrated by the video sequence to be tested comprises:
coding the video sequence to be detected by adopting default quantization parameters in a reference coding mode;
if the code rate after coding is larger than the code rate test upper limit calibrated by the video sequence to be tested, updating the default quantization parameter, continuing to code the video sequence to be tested until the code rate after coding is smaller than or equal to the code rate test upper limit, and taking the latest default quantization parameter as the minimum test point in the first test point set;
and determining other test points in the first test point set based on the minimum test point and a preset quantization point rule.
5. The method of claim 4, wherein the initial code rate in the code rate test segment in which the first test point set is located is a coded code rate for coding the video sequence to be tested by using a maximum test point in the first test point set, and the end code rate is a coded code rate for coding the video sequence to be tested by using a minimum test point in the first test point set.
6. The method of claim 3, wherein if the test points in the first set of test points are quantization points, the determining the second set of test points in the parity coding mode according to the code rate test segment in which the first set of test points is located comprises:
coding the video sequence to be detected by adopting a default initial quantization point in a mark-pair coding mode;
if the code rate after coding exceeds the close range of the termination code rate in the code rate test section where the first test point set is located, updating the default initial quantization point, continuing coding the video sequence to be tested until the code rate after coding is in the close range of the termination code rate, and taking the latest default initial quantization point as the minimum test point in the second test point set;
coding the video sequence to be detected by adopting a default termination quantization point in a label coding mode;
if the code rate after coding exceeds the adjacent range of the initial code rate in the code rate test section where the first test point set is located, updating the default termination quantization point, continuing coding the video sequence to be tested until the code rate after coding is in the adjacent range of the initial code rate, and taking the latest default termination quantization point as the maximum test point in the second test point set;
and determining other test points in the second test point set based on the minimum test point and the maximum test point in the second test point set.
7. The method of claim 6, further comprising, prior to updating the default starting quantization point:
determining the current updating coefficient of the default initial quantization point according to the code rate difference between the coded code rate and the termination code rate;
correspondingly, before updating the default termination quantization point, the method further includes:
and determining the current updating coefficient of the default termination quantization point according to the code rate difference between the coded code rate and the initial code rate.
8. The method of claim 3, wherein the second set of test points is the same as the first set of test points if the test points in the first set of test points are rate points.
9. The method of any of claims 1-8, further comprising, prior to determining the first set of test points in the reference coding mode:
if the video sequence to be tested is in a non-compressed format, the default quantization parameter in a reference coding mode is adopted to code the video sequence to be tested, and the code rate after coding is used as the code rate test upper limit calibrated by the video sequence to be tested, otherwise, the original code rate after the video sequence to be tested is compressed is used as the code rate test upper limit calibrated by the video sequence to be tested.
10. The method of any of claims 1-8, further comprising, prior to determining the first set of test points in the reference coding mode:
and if the video sequence to be detected is in a compressed format, converting the video sequence to be detected into a non-compressed format, and setting the size of the video sequence to be detected according to the lower limit of the coding size.
11. The method of any of claims 1-8, further comprising, prior to determining the first set of test points in the reference coding mode:
and screening out a reference coding mode and a benchmarking coding mode from a coding mode library to be tested.
12. The method according to any of claims 1-8, wherein determining a corresponding coding test result according to a coding performance difference after the video sequence to be tested is coded by using the first test point set in the reference coding mode and the second test point set in the opposite coding mode, respectively, comprises:
coding the video sequence to be tested by respectively adopting the first test point set in the reference coding mode and the second test point set in the label alignment coding mode;
respectively generating Rate Distortion (RD) curves of the video sequence to be detected in the reference coding mode and the benchmarking coding mode by adopting a preset third-party coding evaluation tool, and calculating corresponding coding performance deviation;
and determining a corresponding coding test result based on the coding performance deviation.
13. An apparatus for code testing, comprising:
the first test point determining module is used for determining a first test point set in a reference coding mode according to a code rate test upper limit calibrated by a video sequence to be tested;
the second test point determining module is used for determining a second test point set in a label alignment coding mode according to the code rate test section where the first test point set is located;
and the coding test module is used for determining a corresponding coding test result according to the coding performance difference of the video sequence to be tested after coding by adopting the first test point set in the reference coding mode and the second test point set in the opposite coding mode.
14. A server, characterized in that the server comprises:
one or more processors;
storage means for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement a method of code testing as claimed in any one of claims 1-12.
15. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method of code testing according to any one of claims 1 to 12.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010803701.4A CN111953966B (en) | 2020-08-11 | 2020-08-11 | Method, device, server and storage medium for testing codes |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010803701.4A CN111953966B (en) | 2020-08-11 | 2020-08-11 | Method, device, server and storage medium for testing codes |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111953966A true CN111953966A (en) | 2020-11-17 |
| CN111953966B CN111953966B (en) | 2022-06-28 |
Family
ID=73332163
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010803701.4A Active CN111953966B (en) | 2020-08-11 | 2020-08-11 | Method, device, server and storage medium for testing codes |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111953966B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112995658A (en) * | 2021-04-14 | 2021-06-18 | 浙江华创视讯科技有限公司 | Automatic testing method, testing terminal and system for full performance of coding and decoding |
| CN113099216A (en) * | 2021-03-26 | 2021-07-09 | 北京百度网讯科技有限公司 | Coding complexity evaluation method, device, equipment and storage medium |
| CN113177002A (en) * | 2021-05-24 | 2021-07-27 | 中国工商银行股份有限公司 | Test design method and device based on test points, electronic equipment and medium |
| CN116233438A (en) * | 2023-03-09 | 2023-06-06 | 云贵亮 | Data prediction acquisition system using weighting algorithm |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030179826A1 (en) * | 2002-03-18 | 2003-09-25 | Lg Electronics Inc. | B picture mode determining method and apparatus in video coding system |
| CN101895787A (en) * | 2010-07-21 | 2010-11-24 | 深圳市融创天下科技发展有限公司 | Method and system for subjectively evaluating video coding performance |
| CN102026017A (en) * | 2010-11-29 | 2011-04-20 | 福州瑞芯微电子有限公司 | Efficient test method for video decoding |
| CN103974060A (en) * | 2013-01-31 | 2014-08-06 | 华为技术有限公司 | Method and device for adjusting video quality |
| CN106412572A (en) * | 2016-10-14 | 2017-02-15 | 中国传媒大学 | Video stream encoding quality evaluation method based on motion characteristics |
-
2020
- 2020-08-11 CN CN202010803701.4A patent/CN111953966B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030179826A1 (en) * | 2002-03-18 | 2003-09-25 | Lg Electronics Inc. | B picture mode determining method and apparatus in video coding system |
| CN101895787A (en) * | 2010-07-21 | 2010-11-24 | 深圳市融创天下科技发展有限公司 | Method and system for subjectively evaluating video coding performance |
| CN102026017A (en) * | 2010-11-29 | 2011-04-20 | 福州瑞芯微电子有限公司 | Efficient test method for video decoding |
| CN103974060A (en) * | 2013-01-31 | 2014-08-06 | 华为技术有限公司 | Method and device for adjusting video quality |
| CN106412572A (en) * | 2016-10-14 | 2017-02-15 | 中国传媒大学 | Video stream encoding quality evaluation method based on motion characteristics |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113099216A (en) * | 2021-03-26 | 2021-07-09 | 北京百度网讯科技有限公司 | Coding complexity evaluation method, device, equipment and storage medium |
| CN113099216B (en) * | 2021-03-26 | 2023-03-24 | 北京百度网讯科技有限公司 | Coding complexity evaluation method, device, equipment and storage medium |
| CN112995658A (en) * | 2021-04-14 | 2021-06-18 | 浙江华创视讯科技有限公司 | Automatic testing method, testing terminal and system for full performance of coding and decoding |
| CN112995658B (en) * | 2021-04-14 | 2021-08-03 | 浙江华创视讯科技有限公司 | Automatic testing method, testing terminal and system for full performance of coding and decoding |
| CN113177002A (en) * | 2021-05-24 | 2021-07-27 | 中国工商银行股份有限公司 | Test design method and device based on test points, electronic equipment and medium |
| CN113177002B (en) * | 2021-05-24 | 2024-03-19 | 中国工商银行股份有限公司 | Test design method and device based on test points, electronic equipment and medium |
| CN116233438A (en) * | 2023-03-09 | 2023-06-06 | 云贵亮 | Data prediction acquisition system using weighting algorithm |
| CN116233438B (en) * | 2023-03-09 | 2023-08-29 | 上海华期信息技术有限责任公司 | Data prediction acquisition system using weighting algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111953966B (en) | 2022-06-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111953966B (en) | Method, device, server and storage medium for testing codes | |
| US8102911B2 (en) | Picture coding device | |
| US9030565B2 (en) | Method, apparatus, and system for evaluating video quality | |
| KR101162193B1 (en) | Encoding method, apparatus and device and decoding method | |
| JP2013219822A (en) | Method and apparatus for encoding video, and method and apparatus for decoding video | |
| JP6881819B2 (en) | Video transcoding methods, computer equipment, and storage media | |
| JP2016526336A (en) | System and method for encoding multiple video streams using adaptive quantization for adaptive bitrate streaming | |
| US20080075164A1 (en) | Motion picture encoding apparatus and method | |
| JP2009530892A (en) | Method and apparatus for adapting temporal frequency of video image sequences | |
| CN111726656B (en) | Transcoding method, device, server and storage medium of live video | |
| US20230156238A1 (en) | Heuristic Detection of Potential Digital Media Artifacts and Defects in Digital Media Assets | |
| CN110870312B (en) | Method and apparatus for intra-frame coding in video compression | |
| CN110139168B (en) | Video encoding method, video encoding device, computer equipment and storage medium | |
| Papadopoulos et al. | An adaptive QP offset determination method for HEVC | |
| CN107820084B (en) | Video perception coding method and device | |
| EP2453664A2 (en) | Transcode video verifier device and method for verifying a quality of a transcoded video file | |
| US20200036994A1 (en) | Motion compensation techniques for video | |
| CN111107359A (en) | Intra-frame prediction coding unit dividing method suitable for HEVC standard | |
| WO2023231414A1 (en) | Rate-distortion optimization quantization method and apparatus | |
| JP2019097128A (en) | Video image coding parameter adjustment device, video image coding parameter adjustment method, and program | |
| US10045022B2 (en) | Adaptive content dependent intra prediction mode coding | |
| CN111541898B (en) | Method, device, server and storage medium for determining coding mode | |
| WO2019105535A1 (en) | Device and method for providing an encoded version of a media signal | |
| CN113950842A (en) | Image processing apparatus and method | |
| CN105812803A (en) | Method and device for discarding residual error of TU (transformation unit) |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20221213 Address after: 31a, 15 / F, building 30, maple mall, bangrang Road, Brazil, Singapore Patentee after: Baiguoyuan Technology (Singapore) Co.,Ltd. Address before: 511402 5-13 / F, West Tower, building C, 274 Xingtai Road, Shiqiao street, Panyu District, Guangzhou City, Guangdong Province Patentee before: GUANGZHOU BAIGUOYUAN INFORMATION TECHNOLOGY Co.,Ltd. |
|
| TR01 | Transfer of patent right |