CN111277899B - Video quality evaluation method based on short-term memory and user expectation - Google Patents
Video quality evaluation method based on short-term memory and user expectation Download PDFInfo
- Publication number
- CN111277899B CN111277899B CN202010098992.1A CN202010098992A CN111277899B CN 111277899 B CN111277899 B CN 111277899B CN 202010098992 A CN202010098992 A CN 202010098992A CN 111277899 B CN111277899 B CN 111277899B
- Authority
- CN
- China
- Prior art keywords
- video
- quality
- subjective
- short
- experience
- 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.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/442—Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk
- H04N21/44213—Monitoring of end-user related data
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/47—End-user applications
- H04N21/475—End-user interface for inputting end-user data, e.g. personal identification number [PIN], preference data
- H04N21/4756—End-user interface for inputting end-user data, e.g. personal identification number [PIN], preference data for rating content, e.g. scoring a recommended movie
Abstract
The invention relates to a video quality evaluation method based on short-term memory and user expectation, which is used for acquiring different durations formed by the short-term memory and user expectation video experience data generated after the short-term memory through subjective experiments. And finally, obtaining an evaluation model of the memory forming duration and the video experience quality expected by the user through data analysis and mathematical modeling. The invention considers the influences of short-term memory and user expectation and video quality and can obtain more accurate video quality evaluation.
Description
Technical Field
The invention belongs to the field of video quality evaluation, and particularly relates to a video quality evaluation method based on short-term memory and user expectation.
Background
With the rapid development of networks, various large online network video platforms such as YouTube, Netflix and the like are more popular with people. Because the online video watching is easily influenced by factors such as network and the like, the video playing quality of the online video is fluctuated, thereby bringing different experience qualities to users. Therefore, in order to provide better service for online video users, video providers need to monitor and evaluate the video quality in real time. Many video quality evaluation methods are based on image quality evaluation methods. But unlike images, which are time-domain, the quality of a video may fluctuate at different times, while the quality of an image is fixed. Therefore, the influence factors on the time domain need to be considered when researching the video real-time QoE. While short-term memory is primarily in the temporal domain.
Existing studies have shown that memory has an impact on the quality of video experience, and some scholars exploit memory effects to improve the accuracy of predicting QoE. The effect of memory effects is to leverage past experience to influence the current QoE. It then requires experience of how long time in the past when predicting the current QoE, which we are looking at. When the memory effect is considered in the existing research, the optimal duration of the memory effect is obtained according to a mathematical model. However, at present, human memory cannot be completely described by a mathematical model, and errors necessarily exist.
Disclosure of Invention
In view of the above, the present invention provides a video quality evaluation method based on short-term memory and user expectations.
In order to achieve the purpose, the invention adopts the following technical scheme:
a video quality evaluation method based on short-term memory and user expectation comprises the following steps:
step S1, representing different durations formed by short-term memory through the early-stage experience videos with different durations, carrying out a first subjective experiment, and acquiring first subjective data;
step S2, preprocessing the acquired first subjective data and analyzing to obtain an experimental result of a first subjective experiment;
step S3: designing and carrying out a second subjective experiment based on the analysis result of the first subjective experiment, and acquiring second subjective data of different user expectations for subsequent video experiences formed after short-term memory influence, wherein the user expectations are represented by the internal quality of the previous video;
and step S4, preprocessing the obtained second subjective data, and constructing a video quality evaluation model based on the short-term memory forming duration and the user expectation through a training model.
Step S5, according to the obtained model, predicting the experience quality of the current video by using the memory forming duration and the user expectation, wherein the memory forming duration is the duration of the previous-period video, and the user expectation is the internal quality of the previous-period video;
further, the first subjective experiment and the second subjective experiment are scored by using 11-grade absolute grade scores of ITU-T P.911.
Further, the first subjective experiment is specifically as follows:
step a 1: dividing a 2min source video into two parts for processing, wherein the first 90s is a previous experience video, and the last 30s is a current experience video; the fixed 30s serving as the current experience video is unchanged, and the duration of the previous experience video is changed into 15s, 30s, 45s, 60s and 90 s; thereby obtaining a group of video sequences of the early-stage experience videos with different durations;
step a 2: setting the resolution of the previous experience video to be 640x360 and the resolution of the current experience video to be 1280x720 to obtain a group of video sequences with different durations and variable quality;
step a 3: constructing an experimental video:
video unaffected by short-term memory: the fixed 30s video of the constructed video is provided for observing the experience quality of experimenters not affected by short-term memory, and the experience quality is compared with the experience quality affected by the short-term memory to observe the influence of the short-term memory;
testing a video: namely, the constructed group of video sequences with different time lengths and varying quality;
step a 4: the experimental environment meets the requirements in the ITU-R bt.500-13 recommendation. And scoring the anchor video and the test video by adopting 11 grade absolute grade scores recommended by ITU-T P.911, calculating the average subjective opinion score corresponding to the anchor video and the average subjective opinion score corresponding to the anchor video before and after the quality transformation in the test video, finishing the subjective evaluation of the video quality, and obtaining a subjective evaluation result.
Further, the step S2 is specifically: and screening the obtained first subjective data, removing abnormal data by calculating a correlation coefficient of the subjective data and the MOS value, and removing the abnormal data when the correlation coefficient is less than 0.7.
Further, the second subjective experiment specifically includes:
step b1: taking a video (anchor video) and a test video (test video) which are obtained in the first subjective experiment and are not influenced by short-term memory under the conditions that the early-stage experience time length is 15s, 30s and 45s as experiment video content of the second subjective experiment; redesigning the quality transformation degree of the video, wherein the quality transformation comprises up-switching, down-switching and different variation strengths;
step b2: the experimental environment and scoring criteria were consistent with step a4 in the first subjective experiment.
Further, the step S4 is specifically:
step c 1: the subjective data is first screened for abnormal data in step S2.
Step c 2: secondly, the obtained experimental data are subjected to model training according to the condition that 80% of the experimental data are used as a training set and 20% of the experimental data are used as a testing set. The correlation coefficient and the root mean square error are used to verify the performance of the model. The mathematical relationship models between the user expected value and the subjective quality are respectively as follows under the condition that the early experience duration is 15s, 30s and 45 s:
Q15＝-a1E+b1q+C1 (1)
Q30＝-a2E+b2q+C2 (2)
Q45＝-a3E+b3q+C3 (3)
the expressions (1), (2) and (3) respectively correspond to mathematical relation models between user expectation and subjective quality under the condition that the early experience duration is 15s, 30s and 45 s; e is the expected value generated by watching the previous video, and is determined by the experience quality of the previous video; q is the video quality of experience that is not affected by expectations, i.e. the video intrinsic quality; the model utilizes SSIM to represent the intrinsic quality of the video;
step c 3: defining the expected influence index R as the ratio of the weight a occupied by the expected influence index R to the weight b occupied by the intrinsic quality of the video; the early experience duration is 15s, 30s and 45s, and the corresponding R is respectively: 0.248, 0.330, 0.360. Obtaining a relational expression of the early-stage experience duration t and the expected influence index R:
R＝0.377-0.353e-0.067t (4)
and (3) combining the formulas (1), (2), (3) and (4) to obtain a mathematical relation model of the expected E of the user and the memory forming time t to the subjective quality:
Q＝-R·E+q+C (5)
wherein the desired influence index R is represented by formula (4) and C is a constant term. According to the model, the experience quality of the current video is predicted by using the duration of the previous video and the user expectation.
Compared with the prior art, the invention has the following beneficial effects:
1. the method and the device provided by the invention can be used for acquiring the short-term memory forming duration which enables the influence to be stable by researching the influence of the short-term memory formed in different durations on the user experience quality, and are beneficial to more accurately predicting the current experience quality by utilizing the quality in the early stage
2. On the basis of the stable time, the influence of different user expectation values formed after the influence of short-term memory on the subsequent video experience quality is researched, the user expectation is quantitatively analyzed, and the video quality evaluation at the current moment is further explored.
Drawings
Fig. 1 is a schematic diagram of subjective experimental video construction according to an embodiment of the present invention.
FIG. 2 is a flow chart of subjective experiments in an embodiment of the invention.
Detailed Description
The invention is further explained below with reference to the drawings and the embodiments.
The invention provides a video quality evaluation method based on short-term memory and user expectation, which comprises the following steps:
step S1, representing different durations formed by short-term memory through the early-stage experience videos with different durations, carrying out a first subjective experiment, and acquiring first subjective data;
step S2, preprocessing the acquired first subjective data and analyzing to obtain an experimental result of a first subjective experiment;
step S3: designing and carrying out a second subjective experiment based on the analysis result of the first subjective experiment, and acquiring second subjective data of different user expectations for subsequent video experiences formed after short-term memory influence, wherein the user expectations are represented by the internal quality of the previous video;
and step S4, preprocessing the obtained second subjective data, and constructing a video quality evaluation model based on the short-term memory forming duration and the user expectation through a training model.
Step S5, according to the obtained model, predicting the experience quality of the current video by using the memory forming duration and the user expectation, wherein the memory forming duration is the duration of the previous-period video, and the user expectation is the internal quality of the previous-period video;
in this embodiment, the first subjective experiment and the second subjective experiment both adopt 11-level absolute grade scores of ITU-T p.911 to perform scoring, calculate corresponding average subjective opinion scores before and after quality transformation in each video scene, complete video quality subjective evaluation, and obtain a subjective evaluation result.
In this embodiment, the first subjective experiment is to investigate the influence of different short term memory formation durations on the quality of video experience. Experiments show different durations formed by short-term memory by designing early-stage experience videos with different durations. Video processing of the experiment mainly comprises: and (4) setting the video duration in the early stage and switching the video quality. The subjective experimental process is as follows:
step a1, construction of an experimental video:
processing a source video: the video is processed by dividing the 2min source video into two parts: 1. the first 90s are the prior experience videos, i.e., the videos that form short-term memory (P-video). The previous experience video P-video is cut into different time lengths according to the experimental purpose: 15s, 30s, 45s, 60s, 90 s. 2. The last 30s as the current experience video (C-video). Since the video quality changes can be found clearly in time at the resolution from 640x360 to 1280x720, the invention selects the set of resolution changes to study the influence of the short-term memory formed by the video with the previous experience on the current video experience quality. The resolution of P-video is thus set to 640x360, with durations set to 15s, 30s, 45s, 60s, 90s, respectively. The resolution of C-video is set to 1280x720 and the duration is fixed to 30 s. The video content is continuous.
Secondly, constructing an experimental video sequence: after the above processing, a test video (T-video) and a video (anchor video) not affected by short-term memory can be constructed.
Video unaffected by short-term memory (anchor video): consisting of C-video, for a period of 30 s. The method is used for observing the experience quality of experimenters not influenced by the short-term memory, and can be used for comparing the experience quality with the experience quality influenced by the short-term memory so as to observe the influence of the short-term memory.
Test video (test video): a video sequence consisting of P-video and C-video is used as a test video, wherein the C-video is used for observing the experience quality after being influenced by short-term memory. Since P-video has 5 durations: the duration of C-video is fixed at 30s for 15s, 30s, 45s, 60s, 90s, so that the duration of T-video is 5: 45s, 60s, 75s, 90s, 120 s.
Step a2, experimental environment and step setup:
the experimental environment meets the requirements in the ITU-R bt.500-13 recommendation. The test equipment was a 27 inch 5k liquid crystal display, and since the source video had only a resolution of 3840x2160, the screen resolution display was tuned to 3840x 2160. The video is played using the play software potlayer. The test subjects were 25 persons in total, of which 12 men and 13 women were laymen. The age is 20-25 years, and the vision is normal (including after correction). The experiments were scored for anchor video and test video using 11 scale absolute rating scores as recommended by ITU-T P.911. The experimental procedure was as follows:
firstly, experimenters watch original 4k videos and the video with the worst quality.
Next, all (anchors video) are scored again. The experimental video sequence is played randomly.
And thirdly, scoring the T-video. Before the experimenter scores, the experimenter is explained with a scoring rule that a recorder asks for the experience quality at the current moment at any time during the watching process or directly reports the score when the experimenter feels that the video quality is changed. The experimenter reports the scores in the process of watching the video without interrupting the video watching. Each experimental video sequence was played randomly.
Step a 3: and calculating the average subjective opinion score corresponding to anchor video and the average subjective opinion score corresponding to the test video before and after the quality transformation, and finishing the video quality subjective evaluation to obtain a subjective evaluation result.
In an embodiment of the invention, in the step S2, the result of the first subjective experiment is analyzed and concluded. The subjective data obtained first needs to be screened. And removing abnormal data by calculating a correlation coefficient of the subjective data and the MOS value. This abnormal data is removed when the correlation coefficient is less than 0.7. The correlation coefficient adopts: PLCC, KRCC, SRCC. Secondly, the video score affected by the short-term memory and the video score not affected by the short-term memory are differentiated to obtain a delta MOS, and the result can be obtained by comparing the values of the delta MOS under the condition that the experience time length in the early stage is 15s, 30s, 45s, 60s and 90 s: the longer the duration of short term memory formation the greater the effect and at 45s the effect tends to stabilize.
In this embodiment, in the step S3, on the basis that the stable duration obtained in the first subjective experiment is 45S, and the experience duration in the early stage (i.e. the short-term memory forming duration) of the second subjective experiment is 15S, 30S, and 45S, the influence of different user expectations formed after the short-term memory influence on the quality of experience of the subsequent video is expected. The method specifically comprises the following steps:
step b1, constructing a video:
taking the video and the test video which are obtained in the first subjective experiment and are not affected by the short-term memory under the conditions that the early-stage experience duration is 15s, 30s and 45s as the experiment video content of the second subjective experiment; redesigning the quality transformation degree of the video, wherein the quality transformation comprises up-switching, down-switching and different variation strengths;
step b2: the experimental environment and procedure were consistent with the experimental procedures a2, a3 of the first subjective experiment.
TABLE 1 video quality transition modes
In this embodiment, the step S4 specifically includes:
step c 1: the subjective data is first screened for abnormal data in step S2.
Step c 2: secondly, the obtained experimental data are subjected to model training according to the condition that 80% of the experimental data are used as a training set and 20% of the experimental data are used as a testing set. The correlation coefficient and the root mean square error are used to verify the performance of the model. The mathematical relationship models between the user expected value and the subjective quality are respectively as follows under the condition that the early experience duration is 15s, 30s and 45 s:
Q15＝-a1E+b1q+C1 (1)
Q30＝-a2E+b2q+C2 (2)
Q45＝-a3E+b3q+C3 (3)
the expressions (1), (2) and (3) respectively correspond to mathematical relation models between user expectation and subjective quality under the condition that the early experience duration is 15s, 30s and 45 s; e is the expected value resulting from viewing the previous video, determined by the inherent quality of the previous video; q is the video quality of experience that is not affected by expectations, i.e. the video intrinsic quality; the model utilizes SSIM to represent the intrinsic quality of the video;
step c 3: defining the expected influence index R as the ratio of the weight a occupied by the expected influence index R to the weight b occupied by the intrinsic quality of the video; the early experience duration is 15s, 30s and 45s, and the corresponding R is respectively: 0.248, 0.330, 0.360. Obtaining a relational expression of the early-stage experience duration t and the expected influence index R:
R＝0.377-0.353e-0.067t (4)
and (3) combining the formulas (1), (2), (3) and (4) to obtain a mathematical relation model of the expected E of the user and the memory forming time t to the subjective quality:
Q＝-R·E+q+C (5)
wherein the desired influence index R is represented by formula (4) and C is a constant term.
The above description is only a preferred embodiment of the present invention, and all equivalent changes and modifications made in accordance with the claims of the present invention should be covered by the present invention.
Claims (3)
1. A video quality evaluation method based on short-term memory and user expectations is characterized by comprising the following steps of:
step S1, representing different durations formed by short-term memory through the early-stage experience videos with different durations, carrying out a first subjective experiment, and acquiring first subjective data;
step S2, preprocessing the acquired first subjective data and analyzing to obtain an experimental result of a first subjective experiment;
step S3: designing and carrying out a second subjective experiment based on the analysis result of the first subjective experiment, and acquiring second subjective data of different user expectations for subsequent video experiences formed after short-term memory influence, wherein the user expectations are represented by the internal quality of the previous video;
step S4, preprocessing the obtained second subjective data, and constructing a video quality evaluation model based on short-term memory forming duration and user expectation through a training model;
step S5, according to the obtained model, predicting the experience quality of the current video by using the memory forming duration and the user expectation; the first subjective experiment was specifically as follows:
step a 1: dividing a 2min source video into two parts for processing, wherein the first 90s is a previous experience video, and the last 30s is a current experience video; the fixed 30s serving as the current experience video is unchanged, and the duration of the previous experience video is changed into 15s, 30s, 45s, 60s and 90 s; thereby obtaining a group of video sequences of the early-stage experience videos with different durations;
step a 2: setting the resolution of the previous experience video to be 640x360 and the resolution of the current experience video to be 1280x720 to obtain a group of video sequences with different durations and variable quality;
step a 3: constructing an experimental video:
video unaffected by short-term memory: the fixed 30s video of the constructed video is provided for observing the experience quality of experimenters not affected by short-term memory, and the experience quality is compared with the experience quality affected by the short-term memory to observe the influence of the short-term memory;
testing a video: namely, the constructed group of video sequences with different time lengths and varying quality;
step a 4: grading the video which is not influenced by short-term memory and the test video by adopting 11-grade absolute grade grades recommended by ITU-T P.911, calculating the average subjective opinion score corresponding to the video which is not influenced by the short-term memory and the average subjective opinion score corresponding to the video before and after the quality transformation in the test video, and finishing the subjective evaluation of the video quality to obtain a subjective evaluation result;
the second subjective experiment specifically comprises the following steps:
step b1, constructing an experimental video:
taking the video and the test video which are obtained in the first subjective experiment and are not affected by the short-term memory under the conditions that the early-stage experience duration is 15s, 30s and 45s as the experiment video content of the second subjective experiment; redesigning the quality transformation degree of the video, wherein the quality transformation comprises up-switching, down-switching and different variation strengths;
and b2, adopting 11 grade absolute grade scores recommended by ITU-T P.911 to score videos and test videos which are not affected by short-term memory, calculating the average subjective opinion score corresponding to the videos which are not affected by the short-term memory and the average subjective opinion score corresponding to the videos before and after the quality transformation in the test videos, and finishing the subjective evaluation of the video quality to obtain a subjective evaluation result.
2. The method for evaluating video quality based on short-term memory and user' S desire according to claim 1, wherein the step S2 is specifically as follows: and screening the obtained first subjective data, removing abnormal data by calculating a correlation coefficient of the subjective data and the MOS value, removing the abnormal data when the correlation coefficient is less than 0.7, and analyzing the processed data to obtain a result.
3. The method for evaluating video quality based on short-term memory and user' S desire according to claim 1, wherein the step S4 is specifically as follows:
step c 1: abnormal data screening is carried out on the subjective data;
step c 2: training the obtained experimental data into a model according to the condition that 80% of the obtained experimental data are training sets and 20% of the obtained experimental data are testing sets, verifying the performance of the model by using correlation coefficients and root mean square errors, and obtaining mathematical relation models between user expected values and subjective quality under the condition that the early-stage experience duration is 15s, 30s and 45s, wherein the mathematical relation models are respectively as follows:
the expressions (1), (2) and (3) respectively correspond to mathematical relation models between user expectation and subjective quality under the condition that the early experience duration is 15s, 30s and 45 s; e is the expected value resulting from viewing the previous video, determined by the inherent quality of the previous video; q is the video quality of experience that is not affected by expectations, i.e. the intrinsic quality of the current video;
step c 3: defining the expected influence index R as the ratio of the weight a occupied by the expected influence index R to the weight b occupied by the intrinsic quality of the video; the early experience duration is 15s, 30s and 45s, and the corresponding R is respectively: 0.248, 0.330, 0.360; obtaining a relational expression of the early-stage experience duration t and the expected influence index R:
and (3) combining the formulas (1), (2), (3) and (4) to obtain a mathematical relation model of the expected E of the user and the memory forming time t to the subjective quality:
wherein the desired influence index R is represented by formula (4) and C is a constant term.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010098992.1A CN111277899B (en) | 2020-02-18 | 2020-02-18 | Video quality evaluation method based on short-term memory and user expectation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010098992.1A CN111277899B (en) | 2020-02-18 | 2020-02-18 | Video quality evaluation method based on short-term memory and user expectation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111277899A CN111277899A (en) | 2020-06-12 |
| CN111277899B true CN111277899B (en) | 2021-06-22 |
Family
ID=71002214
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010098992.1A Active CN111277899B (en) | 2020-02-18 | 2020-02-18 | Video quality evaluation method based on short-term memory and user expectation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111277899B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112101788B (en) * | 2020-09-15 | 2022-04-12 | 福州大学 | Video QoE evaluation system and method based on long-term memory |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104796690A (en) * | 2015-04-17 | 2015-07-22 | 浙江理工大学 | Human brain memory model based non-reference video quality evaluation method |
| CN108235001A (en) * | 2018-01-29 | 2018-06-29 | 上海海洋大学 | A kind of deep-sea video quality objective assessment model based on space-time characteristic |
| CN110502665A (en) * | 2019-08-27 | 2019-11-26 | 北京百度网讯科技有限公司 | Method for processing video frequency and device |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101404778B (en) * | 2008-07-16 | 2011-01-19 | 河北师范大学 | Integrated non-reference video quality appraisement method |
| US8917932B2 (en) * | 2011-12-29 | 2014-12-23 | Tektronix, Inc. | Method of predicting perceptual differences between memory and directly viewed colors in video |
| US20150243325A1 (en) * | 2014-02-24 | 2015-08-27 | Lyve Minds, Inc. | Automatic generation of compilation videos |
| US10834406B2 (en) * | 2016-12-12 | 2020-11-10 | Netflix, Inc. | Device-consistent techniques for predicting absolute perceptual video quality |
| US10382829B1 (en) * | 2018-03-23 | 2019-08-13 | Rovi Guides, Inc. | Systems and methods for prompting a user to view an important event in a media asset when the user is using another application |
| CN110366043B (en) * | 2019-08-20 | 2022-02-18 | 北京字节跳动网络技术有限公司 | Video processing method and device, electronic equipment and readable medium |
-
2020
- 2020-02-18 CN CN202010098992.1A patent/CN111277899B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104796690A (en) * | 2015-04-17 | 2015-07-22 | 浙江理工大学 | Human brain memory model based non-reference video quality evaluation method |
| CN108235001A (en) * | 2018-01-29 | 2018-06-29 | 上海海洋大学 | A kind of deep-sea video quality objective assessment model based on space-time characteristic |
| CN110502665A (en) * | 2019-08-27 | 2019-11-26 | 北京百度网讯科技有限公司 | Method for processing video frequency and device |
Non-Patent Citations (2)
| Title |
|---|
| PeQASO: Perceptual Quality Assessment of Streamed Videos Using Optical Flow Features;Mohammed A. Aabed等;《IEEE Transactions on Broadcasting》;20181204;第65卷(第3期);534-545 * |
| 基于记忆效应的HAS移动流媒体QoE评价研究;于世朋;《中国优秀硕士学位论文全文数据库 信息科技辑》;20180415(第4期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111277899A (en) | 2020-06-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ghadiyaram et al. | A subjective and objective study of stalling events in mobile streaming videos | |
| Sinno et al. | Large-scale study of perceptual video quality | |
| Liu et al. | Deriving and validating user experience model for DASH video streaming | |
| Zhang et al. | Subjective and objective quality assessment of panoramic videos in virtual reality environments | |
| CN107027023B (en) | Based on the VoIP of neural network without reference video communication quality method for objectively evaluating | |
| Rodríguez et al. | Video quality assessment in video streaming services considering user preference for video content | |
| CN100559880C (en) | A kind of highly-clear video image quality evaluation method and device based on self-adapted ST area | |
| CN108428227A (en) | Non-reference picture quality appraisement method based on full convolutional neural networks | |
| CN106604078B (en) | A kind of network video recommended method and device | |
| CN109451303A (en) | A kind of modeling method for user experience quality QoE in VR video | |
| CN111277899B (en) | Video quality evaluation method based on short-term memory and user expectation | |
| CN108055533B (en) | Subjective quality evaluation method for panoramic video | |
| Yim et al. | Subjective quality assessment for youtube ugc dataset | |
| Singla et al. | Assessing media qoe, simulator sickness and presence for omnidirectional videos with different test protocols | |
| Corriveau | Video quality testing | |
| CN111479108B (en) | Video and audio joint quality evaluation method and device based on neural network | |
| CN112101788B (en) | Video QoE evaluation system and method based on long-term memory | |
| WO2018153161A1 (en) | Video quality evaluation method, apparatus and device, and storage medium | |
| Chen et al. | QoE evaluation for live broadcasting video | |
| Menkovski et al. | Tackling the sheer scale of subjective QoE | |
| Zhang et al. | Qoecenter: A visual platform for qoe evaluation of streaming video services | |
| Martinez et al. | A no-reference autoencoder video quality metric | |
| Palhais et al. | Quality of experience assessment in internet tv | |
| Salvador et al. | QoE assessment of HTTP adaptive video streaming | |
| Song et al. | Initial perceived quality analysis for dash video streaming |
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 |