WO2020233536A1 - Vr视频质量评估方法及装置 - Google Patents

Vr视频质量评估方法及装置 Download PDF

Info

Publication number
WO2020233536A1
WO2020233536A1 PCT/CN2020/090724 CN2020090724W WO2020233536A1 WO 2020233536 A1 WO2020233536 A1 WO 2020233536A1 CN 2020090724 W CN2020090724 W CN 2020090724W WO 2020233536 A1 WO2020233536 A1 WO 2020233536A1
Authority
WO
WIPO (PCT)
Prior art keywords
video
user
head rotation
log
angle
Prior art date
Application number
PCT/CN2020/090724
Other languages
English (en)
French (fr)
Inventor
熊婕
黄一宏
陈�光
陈建
Original Assignee
华为技术有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to KR1020217039548A priority Critical patent/KR102600721B1/ko
Priority to EP20810116.2A priority patent/EP3958558A4/en
Priority to JP2021566118A priority patent/JP7327838B2/ja
Publication of WO2020233536A1 publication Critical patent/WO2020233536A1/zh
Priority to US17/527,604 priority patent/US20220078447A1/en

Links

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
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N17/00—Diagnosis, testing or measuring for television systems or their details
    • H04N17/02—Diagnosis, testing or measuring for television systems or their details for colour television signals
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T19/00—Manipulating 3D models or images for computer graphics
    • G06T19/20—Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
    • H04N13/10—Processing, recording or transmission of stereoscopic or multi-view image signals
    • H04N13/106—Processing image signals
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N17/00—Diagnosis, testing or measuring for television systems or their details
    • H04N17/004—Diagnosis, testing or measuring for television systems or their details for digital television systems
    • 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]
    • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
    • H04N21/234—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs
    • H04N21/23418—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics
    • 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
    • H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs
    • H04N21/44008—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
    • 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]
    • H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/81—Monomedia components thereof
    • H04N21/816—Monomedia components thereof involving special video data, e.g 3D video
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
    • H04N2013/0074—Stereoscopic image analysis

Definitions

  • the present invention relates to the field of video processing, in particular to a method and device for evaluating VR video quality.
  • VR technology is a cutting-edge technology that combines multiple fields (including computer graphics, human-computer interaction technology, sensor technology, human-machine interface technology, artificial intelligence technology, etc.). By using appropriate equipment to deceive the human senses (such as three-dimensional vision, hearing, smell, etc.), create a world completely separated from reality, and experience and interact with it. To put it simply, VR technology is a technology that uses computers to create a fake world and create an immersive and interactive audio-visual experience. With the increasing popularity of VR services, the VR industry ecosystem has begun. Whether operators, industry partners, or ordinary consumers, VR service quality assessment methods are needed to evaluate user experience, mainly by assessing the quality of VR videos. Evaluate the user's experience, so as to drive the transformation of VR services from usable to easy to use, and achieve the effect of promoting the development of the VR industry.
  • the video quality is evaluated by the bit rate, resolution, and frame rate of the video. This is the evaluation method of traditional video quality.
  • VR video is very different from traditional video.
  • VR video is a 360-degree panoramic video, and VR video has a unique encoding method. If traditional video quality evaluation methods are used to evaluate the quality of VR videos, the problem of low accuracy of the evaluated results will arise.
  • the embodiments of the present invention provide a method and device for evaluating VR video quality, and the adoption of the embodiments of the present invention is beneficial to improve the accuracy of the VR video evaluation result.
  • an embodiment of the present invention provides a VR video quality evaluation method, including:
  • TI temporal perceptual information
  • the TI of the VR video is used to characterize the change in time of the VR video video sequence; according to the VR video bit rate , Frame rate, resolution, and TI determine the mean opinion score MOS of the VR video.
  • the MOS of the VR video is used to characterize the quality of the VR video.
  • the introduction of TI as a parameter for evaluating the quality of VR video improves the accuracy of the evaluation result of the VR video quality.
  • obtaining the TI of the VR video includes:
  • W and H are the width and height of two adjacent frames respectively.
  • obtaining the TI of the VR video includes:
  • acquiring the rotation angle ⁇ a of the user's head within the preset time length ⁇ t includes:
  • ⁇ a (180-abs(y t )+180-abs(y t+ ⁇ t ))-1
  • determining the TI of the VR video according to the user's average head rotation angle includes:
  • the average head rotation angle, m and n are constants.
  • determining the TI of the VR video according to the user's average head rotation angle includes:
  • the average head rotation angle of the user is input into the second TI prediction model for calculation to obtain the TI of the VR video; wherein the second TI prediction model is a non-parametric model.
  • the average subjective opinion score MOS of the VR video is determined according to the bit rate, frame rate, resolution and TI of the VR video, including:
  • MOS 5-a*log(max(log(B1),0.01))-b*log(max(log(B2),0.01))-c*log(max(logF,0.01))-d*log( max(log(TI),0.01));
  • B1 is the bit rate of the VR video,
  • B2 is the resolution of the VR video,
  • F is the frame rate of the VR video, and
  • a, b, c, and d are constants.
  • an embodiment of the present invention provides an evaluation device, including:
  • the acquiring unit is used to acquire the bit rate, frame rate, resolution, and time perception information TI of the VR video, where the TI of the VR video is used to characterize the change in time of the video sequence of the VR video;
  • the determining unit is used to determine the average subjective opinion score MOS of the VR video according to the bit rate, frame rate, resolution and TI of the VR video, where the MOS of the VR video is used to characterize the quality of the VR video.
  • the acquiring unit is configured to:
  • W and H are the width and height of two adjacent frames respectively.
  • the acquiring unit is configured to:
  • the acquiring unit in terms of acquiring the angle ⁇ a of the user's head rotation within the preset time length ⁇ t, the acquiring unit is specifically configured to:
  • ⁇ a (180-abs(y t )+180-abs(y t+ ⁇ t ))-1
  • the acquiring unit is specifically configured to:
  • the acquiring unit is specifically configured to:
  • the average head rotation angle of the user is input into the second TI prediction model for calculation to obtain the TI of the VR video; wherein the second TI prediction model is a non-parametric model.
  • the determining unit is specifically used for:
  • MOS 5-a*log(max(log(B1),0.01))-b*log(max(log(B2),0.01))-c*log(max(logF,0.01))-d*log( max(log(TI),0.01));
  • B1 is the bit rate of the VR video
  • B2 is the resolution of the VR video
  • F is the frame rate of the VR video
  • a, b, c, and d are constants.
  • an embodiment of the present invention provides an evaluation device, including:
  • a memory storing executable program codes; a processor coupled with the memory; the processor calls the executable program codes stored in the memory to execute part or all of the method described in the first aspect.
  • an embodiment of the present invention provides a computer-readable storage medium, the computer storage medium stores a computer program, the computer program includes program instructions, when the program instructions are executed by a processor, the processor executes Part or all of the method described in the first aspect.
  • the bit rate, frame rate, resolution and TI of the VR video are obtained, and the average subjective opinion of the VR video is determined according to the bit rate, frame rate, resolution and TI of the VR video
  • the MOS is divided into MOS, where the MOS of the VR video is used to characterize the quality of the VR video. Using the embodiments of the present invention can help improve the accuracy of the VR video evaluation result.
  • FIG. 1 is a schematic diagram of a scene of VR video quality evaluation provided by an embodiment of the present invention
  • FIG. 2 is a schematic flowchart of a VR video quality evaluation method provided by an embodiment of the present invention
  • FIG. 3 is a schematic structural diagram of an evaluation device provided by an embodiment of the present invention.
  • Figure 4 is a schematic structural diagram of another evaluation device provided by an embodiment of the present invention.
  • FIG. 1 is a schematic diagram of a VR video quality evaluation scene provided by an embodiment of the present invention. As shown in FIG. 1, this scenario includes: a video server 101, an intermediate network device 102, and a terminal device 103.
  • the video server 101 is a server that provides video services for operators and the like.
  • the intermediate network device 102 is a device that implements video transmission between the video server 101 and the terminal device 103, such as a home gateway.
  • the home gateway has not only become the hub connecting the inside and outside, but also the most important control center of the entire home network. It provides a high-speed access interface on the network side for access to the WAN. Provide an Ethernet interface and/or provide wireless local area network functions on the user side to connect various business terminals in the home, such as personal computers and IP set-top boxes.
  • the terminal device 103 is also called User Equipment (UE), which is a device that provides users with voice and/or data connectivity, such as mobile phones, tablet computers, notebook computers, handheld computers, and mobile Internet devices (mobile internet devices).
  • UE User Equipment
  • MID mobile Internet devices
  • wearable devices such as head-mounted devices.
  • any one of the video server 101, the intermediate network device 102 and the terminal device 101 can perform the method for evaluating the quality of the VR video of the present invention.
  • FIG. 2 is a schematic flowchart of a method for evaluating VR video quality according to an embodiment of the present invention. As shown in Figure 2, the method includes:
  • the evaluation device obtains the bit rate, resolution, frame rate and TI of the VR video.
  • the bit rate of the VR video is the transmission rate of the VR video stream per unit time
  • the resolution of the VR video is the resolution of each frame of the VR video
  • the frame rate of the VR video is the number of frames of the refreshed image per unit time.
  • the TI of the VR video is used to indicate the amount of change in the video sequence of the VR video over time. The greater the change in time of the video sequence, the greater the TI value. Generally, a video sequence with a higher degree of motion has a larger amount of change in time, so its TI value is usually larger.
  • the evaluation device calculates the bitstream of the VR video by obtaining the load size of the VR video bitstream in a period of time.
  • the evaluation device obtains the sequence parameter set (SPS) and picture parameter set (PPS) of the VR video by analyzing the bit stream of the VR video, and then determines the resolution of the VR video according to the syntax elements in the SPS and PPS Rate and frame rate.
  • SPS sequence parameter set
  • PPS picture parameter set
  • the evaluation device acquiring the TI of the VR video includes:
  • the evaluation device determines the TI of the VR video according to the pixel values of two adjacent frames of the VR video, including:
  • the evaluation device obtains the difference of pixels at the same position in two adjacent frames of images; calculates the difference between the pixel values of pixels at the same position in two adjacent frames of images according to the standard deviation formula to obtain the TI of the VR video.
  • W and H are the width and height of two adjacent frames respectively.
  • W*H is the resolution of two adjacent frames.
  • the evaluation device determines the TI of the VR video based on the pixel values of the pixels of the consecutive N frames of the VR video
  • the evaluation device determines the VR video based on the pixel values of the pixels of the two adjacent frames of the image described above.
  • the related description of the TI process obtains N-1 candidate TIs, and then the average value of the N-1 candidate TIs is determined as the TI of the VR video, where N is an integer greater than 2.
  • the evaluation device determines the TI of the VR video according to the user's head rotation angle information, including:
  • obtaining the angle ⁇ a of the user's head rotation within the preset time period ⁇ t by the evaluation device includes:
  • ⁇ a (180-abs(y t )+180-abs(y t+ ⁇ t ))-1
  • the preset duration may be the duration of playing one frame of the VR video.
  • the evaluation device determines the TI of the VR video according to the user's average head rotation angle, including:
  • the evaluation device inputs the angleVelocity into the first TI prediction model for calculation to obtain the TI of the VR video.
  • m and n can be set through experience, and the value range of m and n can be [-100, 100], and further, the value range of m and n can be specifically [-50, 50].
  • m and n can also be obtained through training, and m and n obtained through training are usually values in the range [-100, 100].
  • the process of training to obtain m and n is the process of training to obtain the TI prediction model.
  • the evaluation device before the angleVelocity is input into the first TI prediction model for calculation, the evaluation device obtains a first training data set containing multiple data items to train the first parameter model to obtain the first TI prediction model.
  • each first data item in the first training data set includes an average head rotation angle and TI.
  • the average head rotation angle is the input data of the first parameter model
  • TI is the output data of the first parameter model.
  • the first parameter model is a model described by algebraic equations, differential equations, differential equations, and transfer functions.
  • the establishment of the first parameter model is to determine the parameters in the known model structure. For example, m and n in the above TI prediction model.
  • the evaluation device may train the training parameter model through the training data set to obtain parameters in the model, such as m and n in the TI prediction model.
  • the evaluation device determines the TI of the VR video according to the average head rotation angle of the user, including:
  • the second TI prediction model is a non-parametric model.
  • non-parametric models do not make strong assumptions about the form of the objective function. By not making assumptions, they can freely learn any function form from the training data.
  • the training steps of the non-parametric model are similar to those of the parametric model. A large number of training data sets need to be prepared to train the model, but unlike the parametric model that needs to determine an objective function, the non-parametric model does not need to make any changes to the form of the objective function.
  • the KNN k-Nearest Neighbor
  • the evaluation device of the present invention is connected to the user's head mount display (HMD) in a wired or wireless manner, so that the evaluation device can obtain the user's head angle information.
  • HMD head mount display
  • the evaluation device determines the MOS of the VR video according to the bit rate, resolution, frame rate and TI of the VR video.
  • the MOS of VR video is used to characterize the quality of VR video and is an evaluation standard to measure the quality of video.
  • the scoring standard comes from ITU-T P.910.
  • the quality of the video is divided into 5 levels, including: excellent, good, fair, poor, and poor.
  • the corresponding MOSs are 5, 4, 3, 2, and 1, respectively.
  • the evaluation device inputs the bit rate, resolution, frame rate, and TI of the VR video into the quality evaluation model for calculation to obtain the MOS of the VR video.
  • the quality assessment model can be:
  • MOS 5-a*log(max(log(B1),0.01))-b*log(max(log(B2),0.01))-c*log(max(logF,0.01))-d*log( max(log(TI),0.01));
  • B1 is the bit rate of the VR video
  • B2 is the resolution of the VR video
  • F is the frame rate of the VR video
  • a, b, c, and d are constants.
  • a, b, c, and d can be set through experience, and the value range of a, b, c, and d can be [-100, 100], and further, the value range of a, b, c, and d Specifically, it can be [-50,50].
  • a, b, c, and d can also be obtained through training, and the training obtained a, b, c, and d are usually values in the range [-100, 100].
  • the process of training to obtain a, b, c, and d is the process of training to obtain the quality evaluation model.
  • the higher the bit rate of the VR video the greater the MOS value of the VR video, that is, the higher the quality of the VR video; the higher the resolution of the VR video, the greater the MOS value of the VR video; the frame rate of the VR video
  • the higher the value the larger the MOS value of the VR video; the larger the TI value of the VR video, the larger the MOS value of the VR video.
  • the evaluation device before the bit rate, resolution, frame rate, and TI of the VR video are input into the quality evaluation model for calculation, the evaluation device obtains a third training data set containing multiple data items to train the first Two-parameter model to obtain a quality evaluation model.
  • each data item in the third training data set contains information and MOS of a VR video.
  • the information of the VR video is the input data of the second parameter model, and TI is the output data of the second parameter model.
  • the VR video information includes the bit rate, resolution, and frame rate of the VR video.
  • the second parameter model is a model described by algebraic equations, differential equations, differential equations, and transfer functions.
  • the establishment of the second parameter model is to determine each parameter in the known model structure. For example, a, b, c and d in the above quality assessment model.
  • the evaluation device introduces the TI of the VR video to evaluate the quality of the VR video. Compared with the prior art, the accuracy of the quality evaluation of the VR video is significantly improved.
  • the evaluation device 300 includes:
  • the acquiring unit 301 is configured to acquire the bit rate, frame rate, resolution, and time perception information TI of the VR video, where the TI of the VR video is used to characterize the change in time of the video sequence of the VR video;
  • the determining unit 302 is configured to determine the average subjective opinion score MOS of the VR video according to the bit rate, frame rate, resolution and TI of the VR video, where the MOS of the VR video is used to characterize the quality of the VR video.
  • the acquiring unit 301 is configured to:
  • W and H are the width and height of two adjacent frames respectively.
  • the acquiring unit 301 is configured to:
  • the acquiring unit 301 is specifically configured to:
  • ⁇ a (180-abs(y t )+180-abs(y t+ ⁇ t ))-1
  • the acquiring unit 301 is specifically configured to:
  • the acquiring unit 301 is specifically configured to:
  • the average head rotation angle of the user is input into the second TI prediction model for calculation to obtain the TI of the VR video; wherein the second TI prediction model is a non-parametric model.
  • the determining unit 302 is specifically configured to:
  • MOS 5-a*log(max(log(B1),0.01))-b*log(max(log(B2),0.01))-c*log(max(logF,0.01))-d*log( max(log(TI),0.01));
  • B1 is the bit rate of the VR video
  • B2 is the resolution of the VR video
  • F is the frame rate of the VR video
  • a, b, c, and d are constants.
  • the aforementioned units are used to execute the relevant steps of the aforementioned method.
  • the obtaining unit 301 is specifically configured to execute related content of step S201
  • the determining unit 302 is specifically configured to execute related content of step S202.
  • the evaluation device 300 is presented in the form of a unit.
  • the "unit” here can refer to an application-specific integrated circuit (ASIC), a processor and memory that executes one or more software or firmware programs, an integrated logic circuit, and/or other devices that can provide the above functions .
  • ASIC application-specific integrated circuit
  • the above acquisition unit 301 and determination unit 302 may be implemented by the processor 401 of the evaluation device shown in FIG. 4.
  • the evaluation device 400 can be implemented with the structure in FIG. 4.
  • the evaluation device 400 includes at least one processor 401, at least one memory 402 and at least one communication interface 403.
  • the processor 401, the memory 402, and the communication interface 403 are connected through the communication bus and complete mutual communication.
  • the processor 401 may be a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the programs in the above scheme.
  • CPU central processing unit
  • ASIC application-specific integrated circuit
  • the communication interface 403 is used to communicate with other devices or communication networks, such as Ethernet, wireless access network (RAN), wireless local area network (Wireless Local Area Networks, WLAN), etc.
  • devices or communication networks such as Ethernet, wireless access network (RAN), wireless local area network (Wireless Local Area Networks, WLAN), etc.
  • the memory 402 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, random access memory (RAM), or other types that can store information and instructions
  • the dynamic storage device can also be electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, EEPROM), CD-ROM (Compact Disc Read-Only Memory, CD-ROM) or other optical disc storage, optical disc storage (Including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or can be used to carry or store desired program codes in the form of instructions or data structures and can be used by a computer Any other media accessed, but not limited to this.
  • the memory can exist independently and is connected to the processor through a bus.
  • the memory can also be integrated with the processor.
  • the memory 402 is used to store application program codes for executing the above solutions, and the processor 501 controls the execution.
  • the processor 401 is configured to execute application program codes stored in the memory 402.
  • the code stored in the memory 402 can execute related content of the VR video quality evaluation method disclosed in the embodiment shown in FIG. 2. For example: Obtain the bit rate, frame rate, resolution and time perception information TI of VR video, where TI is used to characterize the time variation of the VR video sequence; determine the VR according to the bit rate, frame rate, resolution and TI of the VR video The average subjective opinion of the video is divided into MOS, where the MOS of the VR video is used to characterize the quality of the VR video.
  • An embodiment of the present invention also provides a computer storage medium, wherein the computer storage medium may store a program, and the program includes part or all of the steps of any VR video quality evaluation method recorded in the above method embodiments.
  • the disclosed device may be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of the units is only a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components may be combined or may be Integrate into another system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • the functional units in the various embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the above-mentioned integrated unit can be implemented in the form of hardware or software functional unit.
  • the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable memory.
  • the technical solution of the present invention essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory, A number of instructions are included to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present invention.
  • the aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other various media that can store program codes.
  • the program can be stored in a computer-readable memory, and the memory can include: flash disk , Read-only memory (English: Read-Only Memory, abbreviation: ROM), random access device (English: Random Access Memory, abbreviation: RAM), magnetic disk or optical disc, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Architecture (AREA)
  • Computer Graphics (AREA)
  • Computer Hardware Design (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)

Abstract

本发明公开了一种VR视频质量评估方法，包括：获取VR视频的码率、帧率、分辨率和TI，并根据VR视频的码率、帧率、分辨率和TI确定VR视频的平均主观意见分MOS，其中，VR视频的MOS用于表征VR视频的质量。本发明还公开了一种评估装置。采用本发明实施例可以有利于提高VR视频评估结果的准确性。

Description

VR视频质量评估方法及装置 技术领域
本发明涉及视频处理领域，尤其涉及一种VR视频质量评估方法及装置。
背景技术
虚拟现实(virtual reality，VR)技术是结合多领域(包括计算机图形学、人机交互技术、传感器技术、人机接口技术、人工智能技术等)的前沿技术。通过借助适当装备，欺骗人体感官(比如三维视觉、听觉、嗅觉等)的方式，创造出完全脱离现实的世界，并与其进行体验和交互。简单地说，VR技术就是用计算机创造以假乱真的世界，打造沉浸式、可交互的视听体验的技术。随着VR业务越来越普及，VR产业生态已经起步，无论是运营商、产业伙伴、还是普通消费者，都需要VR业务质量评估方法来评价用户的体验，主要是通过评估VR视频的质量来评价用户的体验，从而牵引VR业务从可用转变到好用，达到促进VR产业的发展的效果。
现有技术中，通过视频的码率、分辨率和帧率来评估视频的质量。这是传统视频的质量的评估方法。然而VR视频与传统视频有很大的差别，VR视频是360度全景视频，VR视频有特有的编码方式。若传统视频的质量评估方法来评估VR视频的质量，则产生评估出来的结果准确性不高的问题。
发明内容
本发明实施例提供一种VR视频质量的评估方法及装置，采用本发明的实施例有利于提高VR视频评估结果的准确性。
第一方面，本发明实施例提供一种VR视频质量评估方法，包括：
获取VR视频的码率、帧率、分辨率和时间感知信息(temporal perceptual information，TI)，其中，VR视频的TI用于表征VR视频的视频序列在时间上变化量；根据VR视频的码率、帧率、分辨率和TI确定VR视频的平均主观意见分(mean opinion score)MOS，其中，VR视频的MOS用于表征VR视频的质量。与现有技术相比，通过引入TI作为评估VR视频质量的参数，提高了VR视频质量评估结果的准确性。
在一个可行的实施例中，获取VR视频的TI，包括：
获取VR视频的相邻两帧图像中相同位置的像素值的差值；根据标准差公式对相邻两帧图像中相同位置的像素值的差值进行计算，以得到VR视频的TI；
其中，标准差公式为：
Figure PCTCN2020090724-appb-000001
其中，p ij为相邻两帧图像中当前帧的第i行第j个像素点的像素值与该当前帧的前一帧的第i行第j个像素点的像素值之间的差值，W和H分别为相邻两帧图像的宽度和高度。
在一个可行的实施例中，获取VR视频的TI，包括：
获取在预设时长Δt内用户的头部转动的角度Δa；根据预设时长Δt和用户的头部转动的角度Δa确定用户的平均头部转动角；根据用户的平均头部转动角度确定VR视频的TI，其中， 用户的平均头部转动角度越大，VR视频的TI值越大。
在一个可行的实施例中，获取在预设时长Δt内用户的头部转动的角度Δa，包括：
获取t时刻用户头部的角度y t和t+Δt时刻用户头部的角度y t+Δt；并按照以下方法确定用户头部转动的角度Δa：当y t+Δt与y t差值的绝对值大于180度，且y t小于y t+Δt时，
Δa＝180-abs(y t)+180-abs(y t+Δt)
当y t+Δt与y t差值的绝对值大于180度，且y t大于y t+Δt时，
Δa＝(180-abs(y t)+180-abs(y t+Δt))-1
当y t+Δt与y t差值的绝对值不大于180度时，Δa＝y t+Δt-y t。
在一个可行的实施例中，根据用户的平均头部转动角度确定VR视频的TI，包括：
将用户的平均头部转动的角度输入到第一TI预测模型中进行计算，以得到VR视频的TI；其中，第一TI预测模型为：TI＝log(m*angleVelocity)+n；angleVelocity为用户的平均头部转动角度，m和n为常数。通过基于用户头部转动的角度来预测VR视频的TI，使得计算TI所需的计算能力可以忽略不计。
在一个可行的实施例中，根据用户的平均头部转动角度确定VR视频的TI，包括：
将用户的平均头部转动的角度输入到第二TI预测模型中进行计算，以得到VR视频的TI；其中，第二TI预测模型为非参数模型。通过基于用户头部转动的角度来预测VR视频的TI，使得计算TI所需的计算能力可以忽略不计。
在一个可行的实施例中，根据VR视频的码率、帧率、分辨率和TI确定VR视频的平均主观意见分MOS，包括：
将VR视频的码率、分辨率、帧率和TI输入到质量评估模型中进行计算，以得到VR视频的MOS；其中，质量评估模型为：
MOS＝5-a*log(max(log(B1),0.01))-b*log(max(log(B2),0.01))-c*log(max(logF,0.01))-d*log(max(log(TI),0.01))；B1为VR视频的码率，B2为VR视频的分辨率，F为VR视频的帧率，a,b,c,d为常数。
第二方面，本发明实施例提供一种评估装置，包括：
获取单元，用于获取VR视频的码率、帧率、分辨率和时间感知信息TI，其中，VR视频的TI用于表征VR视频的视频序列在时间上的变化量；
确定单元，用于根据VR视频的码率、帧率、分辨率和TI确定VR视频的平均主观意见分MOS，其中，VR视频的MOS用于表征VR视频的质量。
在一个可行的实施例中，在获取VR视频的TI的方面，获取单元具有用于：
获取VR视频的相邻两帧图像中相同位置的像素值的差值；根据标准差公式对相邻两帧图像中相同位置的像素值的差值进行计算，以得到VR视频的TI；
其中，标准差公式为：
Figure PCTCN2020090724-appb-000002
其中，p ij为相邻两帧图像中当前帧的第i行第j个像素点的像素值与该当前帧的前一帧的第i行第j个像素点的像素值之间的差值，W和H分别为相邻两帧图像的宽度和高度。
在一个可行的实施例中，在获取VR视频的TI的方面，获取单元具有用于：
获取在预设时长Δt内用户的头部转动的角度Δa；根据预设时长Δt和用户的头部转动的角度Δa确定用户的平均头部转动角；根据用户的平均头部转动角度确定VR视频的TI，其中，用户的平均头部转动角度越大，VR视频的TI值越大。
在一个可行的实施例中，在获取在预设时长Δt内用户的头部转动的角度Δa的方面，获取单元具体用于：
获取t时刻用户头部的角度y t和t+Δt时刻用户头部的角度y t+Δt；按照以下方法确定用户头部转动的角度Δa：当y t+Δt与y t差值的绝对值大于180度，且y t小于y t+Δt时，
Δa＝180-abs(y t)+180-abs(y t+Δt)
当y t+Δt与y t差值的绝对值大于180度，且y t大于y t+Δt时，
Δa＝(180-abs(y t)+180-abs(y t+Δt))-1
当y t+Δt与y t差值的绝对值不大于180度时，Δa＝y t+Δt-y t。
在一个可行的实施例中，在根据用户的平均头部转动角度确定VR视频的TI的方面，获取单元具体用于：
将用户的平均头部转动的角度输入到第一TI预测模型中进行计算，以得到VR视频的TI；其中，第一TI预测模型为：TI＝log(m*angleVelocity)+n；angleVelocity为用户的平均头部转动角度，m和n为常数。
在一个可行的实施例中，在根据用户的平均头部转动角度确定VR视频的TI的方面，获取单元具体用于：
将用户的平均头部转动的角度输入到第二TI预测模型中进行计算，以得到VR视频的TI；其中，第二TI预测模型为非参数模型。
在一个可行的实施例中，在根据VR视频的码率、帧率、分辨率和TI确定VR视频的平均主观意见分MOS的方面，确定单元具体用于：
将VR视频的码率、分辨率、帧率和TI输入到质量评估模型中进行计算，以得到VR视频的MOS；其中，质量评估模型为：
MOS＝5-a*log(max(log(B1),0.01))-b*log(max(log(B2),0.01))-c*log(max(logF,0.01))-d*log(max(log(TI),0.01))；
B1为VR视频的码率，B2为VR视频的分辨率，F为VR视频的帧率，a,b,c,d为常数。
第三方面，本发明实施例提供一种评估装置，包括：
存储有可执行程序代码的存储器；与该存储器耦合的处理器；该处理器调用存储器中存储的可执行程序代码，执行如第一方面所述方法的部分或全部。
第四方面，本发明实施例提供一种计算机可读存储介质，该计算机存储介质存储有计算机程序，该计算机程序包括程序指令，当该程序指令当被处理器执行时使所述处理器执行如第一方面所述方法的部分或全部。
可以看出，在本发明实施例的方案中，获取VR视频的码率、帧率、分辨率和TI，并根据VR视频的码率、帧率、分辨率和TI确定VR视频的平均主观意见分MOS，其中，VR视频的MOS用于表征VR视频的质量，采用本发明实施例可以有利于提高VR视频评估结果的准确性。
本发明的这些方面或其他方面在以下实施例的描述中会更加简明易懂。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些 附图获得其他的附图。
图1为本发明实施例提供的一种VR视频质量评估的场景示意图；
图2为本发明实施例提供的一种VR视频质量评估方法的流程示意图；
图3为本发明实施例提供的一种评估装置的结构示意图；
图4为本发明实施例提供的另一种评估装置的结构示意图.
具体实施方式
以下分别进行详细说明。
参见图1，图1为本发明实施例提供的一种VR视频质量评估的场景示意图。如图1所示，该场景包括：视频服务器101、中间网络设备102和终端设备103。
其中，视频服务器101为运营商等提供视频服务的服务器。
中间网络设备102为实现视频服务器101和终端设备103之间的视频传输的设备，比如家庭网关。家庭网关不仅成为衔接内外的枢纽，更是整个家庭网络最重要的控制中心，在网络侧提供高速的接入接口，用于接入广域网。在用户侧提供以太网接口和/或提供无线局域网功能，用于连接家庭内的各种业务终端，如个人计算机、IP机顶盒。
终端设备103又称之为用户设备(User Equipment，UE)，是一种向用户提供语音和/或数据连通性的设备，比如手机、平板电脑、笔记本电脑、掌上电脑、移动互联网设备(mobile internet device，MID)、可穿戴设备，比如头戴式设备等。
在本发明中，视频服务器101、中间网络设备102和终端设备101中的任一个均可执行本发明的VR视频的质量评估的方法。
参见图2，图2为本发明实施例提供的一种VR视频质量评估方法的流程示意图。如图2所示，该方法包括：
S201、评估装置获取VR视频的码率，分辨率、帧率和TI。
其中，VR视频的码率为单位时间内传输VR视频码流的速率，VR视频的分辨率为VR视频中每帧图像的分辨率，VR视频的帧率为单位时间为刷新图像的帧数。VR视频的TI用于表示VR视频的视频序列在时间上的变化量。视频序列在时间上的变化量越大，其TI值也越大。通常，运动程度较高的视频序列在时间上的变化量较大，因此其TI值通常较大。
在一个可行的实施例中，评估装置通过获取一段时间内VR视频比特流的负载大小计算VR视频的码流。评估装置通过解析VR视频的比特流，得到VR视频的序列参数集(sequence parameter set，SPS)和图像参数集(picture parameter set,PPS)，然后根据SPS和PPS中的语法元素确定VR视频的分辨率和帧率。
在一个可行的实施例中，评估装置获取VR视频的TI，包括：
采用ITU-R BT.1788中的方式根据VR视频中相邻两帧图像的像素值确定VR视频的TI，或者；
根据用户头部转动角度信息确定VR视频的TI。
具体地，评估装置根据VR视频中相邻两帧图像的像素值确定VR视频的TI，包括：
评估装置获取相邻两帧图像中相同位置像素点的差值；根据标准差公式对相邻两帧图像中相同位置像素点的像素值的差值进行计算，以得到VR视频的TI。
其中，标准差公式为：
Figure PCTCN2020090724-appb-000003
其中，p ij为相邻两帧图像中当前帧的第i行第j个像素点的像素值与该当前帧的前一帧的第i行第j个像素点的像素值之间的差值，W和H分别为相邻两帧图像的宽度和高度。换句话说，W*H为相邻两帧图像的分辨率。
在一个示例中，若评估装置根据VR视频的连续N帧图像的像素点的像素值确定VR视频的TI，则评估装置根据按照上述根据相邻两帧图像的像素点的像素值确定VR视频的TI过程的相关描述得到N-1个候选TI，然后将该N-1个候选TI的平均值确定为VR视频的TI，其中，N为大于2的整数。
在一个可行的实施例中，评估装置根据用户头部转动角度信息确定VR视频的TI，包括：
获取在预设时长Δt内用户头部转动的角度Δa；
根据预设时长Δt和用户头部转动的角度Δa确定用户的平均头部转动角；
根据用户的平均头部转动角度确定VR视频的TI。
具体地，评估装置获取在预设时长Δt内用户头部转动的角度Δa，包括：
获取t时刻用户头部的角度y t和t+Δt时刻用户头部的角度y t+Δt，然后按照以下方法确定用户头部转动的角度Δa：
当y t+Δt与y t差值的绝对值大于180度，且y t小于y t+Δt时，
Δa＝180-abs(y t)+180-abs(y t+Δt)
当y t+Δt与y t差值的绝对值大于180度，且y t大于y t+Δt时，
Δa＝(180-abs(y t)+180-abs(y t+Δt))-1
当y t+Δt与y t差值的绝对值不大于180度时，Δa＝y t+Δt-y t。
评估装置再根据预设时长Δt和用户头部转动的角度Δa确定用户的平均头部转动的角度angleVelocity，其中，该angleVelocity＝Δa/Δt。
需要指出的是，预设时长可以为播放VR视频的一帧图像的时长。
在一个可能的实施例中，评估装置根据用户的平均头部转动角度确定VR视频的TI，包括：
评估装置将angleVelocity输入到第一TI预测模型中进行计算，以得到VR视频的TI。
需要说明的是，angleVelocity越大，VR视频的TI值越大。
可选地，TI预测模型为：TI＝log(m*angleVelocity)+n。其中，m和n均为常数。
可选地，m和n可通过经验进行设置，且m和n的取值范围可以为[-100,100]，进一步地，m和n的取值范围具体可以为[-50,50]。
可选地，m和n还可通过训练得到，训练得到的m和n通常为范围[-100,100]内的值。训练得到m和n的过程就是训练得到TI预测模型的过程。
在一个可行的实施例中，在将angleVelocity输入到第一TI预测模型中进行计算之前，评估装置获取包含多条数据项的第一训练数据集来训练第一参数模型，以得到第一TI预测模型。其中，第一训练数据集中每条第一数据项包含一个平均头部转动角度和TI。平均头部转动角度为第一参数模型的输入数据，TI为第一参数模型的输出数据。
需要说明的是，第一参数模型是用代数方程、微分方程、微分方程组以及传递函数等描述的模型，建立第一参数模型就是确定已知模型结构中的各个参数。比如上述TI预测模型中的m和n。
在一个示例中，评估装置可通过训练数据集对训练参数模型进行训练，得到模型中的参数，比如TI预测模型中的m和n。
在一个可行的实施例中，评估装置根据用户的平均头部转动角度确定VR视频的TI，包括：
将angleVelocity输入到第二TI预测模型中进行计算，以得到VR视频的TI。其中，第二TI预测模型为非参数模型。
在此需要说明的是，非参数模型不对目标函数的形式作出强烈假设，通过不做假设，它们可以从训练数据中自由地学习任何函数形式。非参数模型的训练步骤与参数模型训练方式类似，同样需要准备大量的训练数据集来训练模型，只是不像在参数模型需要确定一个目标函数，非参模型不需要对目标函数的形式作出任何的假设，如可采用k最邻近算法((k-Nearest Neighbor，KNN)算法。
需要指出的是，本发明的评估装置与用户的头显设备(head mount display，HMD)通过有线或者无线方式连接，使得评估装置能够获取用户的头部角度信息。
S202、评估装置根据VR视频的码率，分辨率、帧率和TI确定VR视频的MOS。
其中，VR视频的MOS用于表征VR视频的质量，是一种衡量视频质量好坏的评价标准，打分标准来自ITU-T P.910。视频的质量分为5个等级，包括：优、良、尚可、差和劣，对应的MOS分别为5、4、3、2和1。
具体地，评估装置将VR视频的码率、分辨率、帧率和TI输入到质量评估模型中进行计算，以得到VR视频的MOS。
可选地，质量评估模型可以为：
MOS＝5-a*log(max(log(B1),0.01))-b*log(max(log(B2),0.01))-c*log(max(logF,0.01))-d*log(max(log(TI),0.01))；
其中，B1为VR视频的码率，B2为VR视频的分辨率，F为VR视频的帧率，a,b,c,d为常数。
可选地，a,b,c和d可通过经验进行设置，且a,b,c和d的取值范围可以为[-100,100]，进一步地，a,b,c和d的取值范围具体可以为[-50,50]。
可选地，a,b,c和d还可通过训练得到，训练得到的a,b,c和d通常为范围[-100,100]内的值。训练得到a,b,c和d的过程就是训练得到质量评估模型的过程。
需要说明的是，VR视频的码率越高，VR视频的MOS值越大，即VR视频的质量越高；VR视频的分辨率越高，VR视频的MOS值越大；VR视频的帧率越高，VR视频的MOS值越大；VR视频的TI值越大，VR视频的MOS值越大。
在一个可行的实施例中，在将VR视频的码率、分辨率、帧率和TI输入到质量评估模型中进行计算之前，评估装置获取包含多条数据项的第三训练数据集来训练第二参数模型，以得到质量评估模型。其中，第三训练数据集中每条数据项包含一个VR视频的信息和MOS。VR视频的信息为第二参数模型的输入数据，TI为第二参数模型的输出数据。其中，VR的视频信息包括VR视频的码率、分辨率和帧率。
需要说明的是，第二参数模型是用代数方程、微分方程、微分方程组以及传递函数等描述的模型，建立第二参数模型就是确定已知模型结构中的各个参数。比如上述质量评估模型中的a,b,c和d。
可以看出，在本发明实施例的方案中，评估装置引入VR视频的TI来评估VR视频的质 量，与现有技术相比，显著提高到了VR视频质量评估的准确性。
参见图3，图3为本发明实施例提供的一种评估装置的结构示意图。如图3所示，该评估装置300包括：
获取单元301，用于获取VR视频的码率、帧率、分辨率和时间感知信息TI，其中，VR视频的TI用于表征VR视频的视频序列在时间上的变化量；
确定单元302，用于根据VR视频的码率、帧率、分辨率和TI确定VR视频的平均主观意见分MOS，其中，VR视频的MOS用于表征VR视频的质量。
在一个可行的实施例中，在获取VR视频的TI的方面，获取单元301具有用于：
获取VR视频的相邻两帧图像中相同位置的像素值的差值；根据标准差公式对相邻两帧图像中相同位置的像素值的差值进行计算，以得到VR视频的TI；
其中，标准差公式为：
Figure PCTCN2020090724-appb-000004
其中，p ij为相邻两帧图像中当前帧的第i行第j个像素点的像素值与该当前帧的前一帧的第i行第j个像素点的像素值之间的差值，W和H分别为相邻两帧图像的宽度和高度。
在一个可行的实施例中，在获取VR视频的TI的方面，获取单元301具有用于：
获取在预设时长Δt内用户的头部转动的角度Δa；根据预设时长Δt和用户的头部转动的角度Δa确定用户的平均头部转动角；根据用户的平均头部转动角度确定VR视频的TI，其中，用户的平均头部转动角度越大，VR视频的TI值越大。
在一个可行的实施例中，在获取在预设时长Δt内用户的头部转动的角度Δa的方面，获取单元301具体用于：
获取t时刻用户头部的角度y t和t+Δt时刻用户头部的角度y t+Δt；按照以下方法确定用户头部转动的角度Δa：当y t+Δt与y t差值的绝对值大于180度，且y t小于y t+Δt时，
Δa＝180-abs(y t)+180-abs(y t+Δt)
当y t+Δt与y t差值的绝对值大于180度，且y t大于y t+Δt时，
Δa＝(180-abs(y t)+180-abs(y t+Δt))-1
当y t+Δt与y t差值的绝对值不大于180度时，Δa＝y t+Δt-y t。
在一个可行的实施例中，在根据用户的平均头部转动角度确定VR视频的TI的方面，获取单元301具体用于：
将用户的平均头部转动的角度输入到第一TI预测模型中进行计算，以得到VR视频的TI；其中，第一TI预测模型为：TI＝log(m*angleVelocity)+n；angleVelocity为用户的平均头部转动角度，m和n为常数。
在一个可行的实施例中，在根据用户的平均头部转动角度确定VR视频的TI的方面，获取单元301具体用于：
将用户的平均头部转动的角度输入到第二TI预测模型中进行计算，以得到VR视频的TI；其中，第二TI预测模型为非参数模型。
在一个可行的实施例中，在根据VR视频的码率、帧率、分辨率和TI确定VR视频的平均主观意见分MOS的方面，确定单元302具体用于：
将VR视频的码率、分辨率、帧率和TI输入到质量评估模型中进行计算，以得到VR视频的MOS；其中，质量评估模型为：
MOS＝5-a*log(max(log(B1),0.01))-b*log(max(log(B2),0.01))-c*log(max(logF,0.01))-d*log(max(log(TI),0.01))；
B1为VR视频的码率，B2为VR视频的分辨率，F为VR视频的帧率，a,b,c,d为常数。
需要说明的是，上述各单元(获取单元301和确定单元302)用于执行上述方法的相关步骤。其中，获取单元301具体用于执行步骤S201的相关内容，确定单元302具体用于执行步骤S202的相关内容。
在本实施例中，评估装置300是以单元的形式来呈现。这里的“单元”可以指特定应用集成电路(application-specific integrated circuit，ASIC)，执行一个或多个软件或固件程序的处理器和存储器，集成逻辑电路，和/或其他可以提供上述功能的器件。此外，以上获取单元301和确定单元302可通过图4所示的评估装置的处理器401来实现。
如图4所示评估装置400可以以图4中的结构来实现，该评估装置400包括至少一个处理器401，至少一个存储器402以及至少一个通信接口403。所述处理器401、所述存储器402和所述通信接口403通过所述通信总线连接并完成相互间的通信。
处理器401可以是通用中央处理器(CPU)，微处理器，特定应用集成电路(application-specific integrated circuit，ASIC)，或一个或多个用于控制以上方案程序执行的集成电路。
通信接口403，用于与其他设备或通信网络通信，如以太网，无线接入网(RAN)，无线局域网(Wireless Local Area Networks，WLAN)等。
存储器402可以是只读存储器(read-only memory，ROM)或可存储静态信息和指令的其他类型的静态存储设备，随机存取存储器(random access memory，RAM)或者可存储信息和指令的其他类型的动态存储设备，也可以是电可擦可编程只读存储器(Electrically Erasable Programmable Read-Only Memory，EEPROM)、只读光盘(Compact Disc Read-Only Memory，CD-ROM)或其他光盘存储、光碟存储(包括压缩光碟、激光碟、光碟、数字通用光碟、蓝光光碟等)、磁盘存储介质或者其他磁存储设备、或者能够用于携带或存储具有指令或数据结构形式的期望的程序代码并能够由计算机存取的任何其他介质，但不限于此。存储器可以是独立存在，通过总线与处理器相连接。存储器也可以和处理器集成在一起。
其中，所述存储器402用于存储执行以上方案的应用程序代码，并由处理器501来控制执行。所述处理器401用于执行所述存储器402中存储的应用程序代码。
存储器402存储的代码可执行图2所示实施例公开的VR视频质量评估方法的相关内容。比如：获取VR视频的码率、帧率、分辨率和时间感知信息TI，其中，TI用于表征VR视频序列的时间变化量；根据VR视频的码率、帧率、分辨率和TI确定VR视频的平均主观意见分MOS，其中，VR视频的MOS用于表征VR视频的质量。
本发明实施例还提供一种计算机存储介质，其中，该计算机存储介质可存储有程序，该程序执行时包括上述方法实施例中记载的任何一种VR视频质量评估方法的部分或全部步骤。
需要说明的是，对于前述的各方法实施例，为了简单描述，故将其都表述为一系列的动作组合，但是本领域技术人员应该知悉，本发明并不受所描述的动作顺序的限制，因为依据本发明，某些步骤可以采用其他顺序或者同时进行。其次，本领域技术人员也应该知悉，说明书中所描述的实施例均属于优选实施例，所涉及的动作和模块并不一定是本发明所必须的。
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述的部分，可 以参见其他实施例的相关描述。
在本申请所提供的几个实施例中，应该理解到，所揭露的装置，可通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本发明各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储器中。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储器中，包括若干指令用以使得一台计算机设备(可为个人计算机、服务器或者网络设备等)执行本发明各个实施例所述方法的全部或部分步骤。而前述的存储器包括：U盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、移动硬盘、磁碟或者光盘等各种可以存储程序代码的介质。
本领域普通技术人员可以理解上述实施例的各种方法中的全部或部分步骤是可以通过程序来指令相关的硬件来完成，该程序可以存储于一计算机可读存储器中，存储器可以包括：闪存盘、只读存储器(英文：Read-Only Memory，简称：ROM)、随机存取器(英文：Random Access Memory，简称：RAM)、磁盘或光盘等。
以上对本发明实施例进行了详细介绍，本文中应用了具体个例对本发明的原理及实施方式进行了阐述，以上实施例的说明只是用于帮助理解本发明的方法及其核心思想；同时，对于本领域的一般技术人员，依据本发明的思想，在具体实施方式及应用范围上均会有改变之处，综上上述，本说明书内容不应理解为对本发明的限制。

Claims (14)

  1. 一种虚拟现实VR视频质量评估方法，其特征在于，包括：
    获取VR视频的码率、帧率、分辨率和时间感知信息TI，其中，所述VR视频的TI用于表征所述VR视频的视频序列在时间上的变化量；
    根据所述VR视频的码率、帧率、分辨率和TI确定所述VR视频的平均主观意见分MOS，其中，所述VR视频的MOS用于表征所述VR视频的质量。
  2. 根据权利要求1所述的方法，其特征在于，所述获取VR视频的TI，包括：
    获取在预设时长Δt内用户的头部转动的角度Δa；
    根据所述预设时长Δt和所述用户的头部转动的角度Δa确定用户的平均头部转动角；
    根据所述用户的平均头部转动角度确定所述VR视频的TI，其中，所述用户的平均头部转动角度越大，所述VR视频的TI值越大。
  3. 根据权利要求2所述的方法，其特征在于，所述获取在预设时长Δt内用户的头部转动的角度Δa，包括：
    获取t时刻用户头部的角度y t和t+Δt时刻用户头部的角度y t+Δt；
    按照以下方法确定用户头部转动的角度Δa：
    当y t+Δt与y t差值的绝对值大于180度，且y t小于y t+Δt时，
    Δa＝180-abs(y t)+180-abs(y t+Δt)
    当y t+Δt与y t差值的绝对值大于180度，且y t大于y t+Δt时，
    Δa＝(180-abs(y t)+180-abs(y t+Δt))-1
    当y t+Δt与y t差值的绝对值不大于180度时，Δa＝y t+Δt-y t。
  4. 根据权利要求2或3所述的方法，其特征在于，所述根据所述用户的平均头部转动角度确定所述VR视频的TI，包括：
    将所述用户的平均头部转动的角度输入到第一TI预测模型中进行计算，以得到所述VR视频的TI；
    其中，所述第一TI预测模型为：TI＝log(m*angleVelocity)+n；
    所述angleVelocity为用户的平均头部转动角度，所述m和n为常数。
  5. 根据权利要求2或3所述的方法，其特征在于，所述根据所述用户的平均头部转动角度确定所述VR视频的TI，包括：
    将所述用户的平均头部转动的角度输入到第二TI预测模型中进行计算，以得到所述VR视频的TI；
    其中，所述第二TI预测模型为非参数模型。
  6. 根据权利要求1-5任一项所述的方法，其特征在于，所述根据所述VR视频的码率、帧率、分辨率和TI确定所述VR视频的平均主观意见分MOS，包括：
    将所述VR视频的码率、分辨率、帧率和TI输入到质量评估模型中进行计算，以得到所述VR视频的MOS；
    其中，所述质量评估模型为：
    MOS＝5-a*log(max(log(B1),0.01))-b*log(max(log(B2),0.01))-c*log(max(logF,0.01))-d*log(max(log(TI),0.01))；
    所述B1为所述VR视频的码率，所述B2为所述VR视频的分辨率，所述F为所述VR视频的帧率，a,b,c,d为常数。
  7. 一种评估装置，其特征在于，包括：
    获取单元，用于获取虚拟现实VR视频的码率、帧率、分辨率和时间感知信息TI，其中，所述VR视频的TI用于表征所述VR视频的视频序列在时间上的变化量；
    确定单元，用于根据所述VR视频的码率、帧率、分辨率和TI确定所述VR视频的平均主观意见分MOS，其中，所述VR视频的MOS用于表征所述VR视频的质量。
  8. 根据权利要求7所述的装置，其特征在于，在所述获取VR视频的TI的方面，获取单元具有用于：
    获取在预设时长Δt内用户的头部转动的角度Δa；
    根据所述预设时长Δt和所述用户的头部转动的角度Δa确定用户的平均头部转动角；
    根据所述用户的平均头部转动角度确定所述VR视频的TI，其中，所述用户的平均头部转动角度越大，所述VR视频的TI值越大。
  9. 根据权利要求8所述的装置，其特征在于，在所述获取在预设时长Δt内用户的头部转动的角度Δa的方面，所述获取单元具体用于：
    获取t时刻用户头部的角度y t和t+Δt时刻用户头部的角度y t+Δt；
    按照以下方法确定用户头部转动的角度Δa：
    当y t+Δt与y t差值的绝对值大于180度，且y t小于y t+Δt时，
    Δa＝180-abs(y t)+180-abs(y t+Δt)
    当y t+Δt与y t差值的绝对值大于180度，且y t大于y t+Δt时，
    Δa＝(180-abs(y t)+180-abs(y t+Δt))-1
    当y t+Δt与y t差值的绝对值不大于180度时，Δa＝y t+Δt-y t。
  10. 根据权利要求8或9所述的装置，其特征在于，在所述根据所述用户的平均头部转动角度确定所述VR视频的TI的方面，所述获取单元具体用于：
    将所述用户的平均头部转动的角度输入到第一TI预测模型中进行计算，以得到所述VR视频的TI；
    其中，所述第一TI预测模型为：TI＝log(m*angleVelocity)+n；
    所述angleVelocity为用户的平均头部转动角度，所述m和n为常数。
  11. 根据权利要求8或9所述的装置，其特征在于，在所述根据所述用户的平均头部转动角度确定所述VR视频的TI的方面，所述获取单元具体用于：
    将所述用户的平均头部转动的角度输入到第二TI预测模型中进行计算，以得到所述VR视频的TI；
    其中，所述第二TI预测模型为非参数模型。
  12. 根据权利要求7-11任一项所述的装置，其特征在于，在所述根据所述VR视频的码率、帧率、分辨率和TI确定所述VR视频的平均主观意见分MOS的方面，所述确定单元具体用于：
    将所述VR视频的码率、分辨率、帧率和TI输入到质量评估模型中进行计算，以得到所述VR视频的MOS；
    其中，所述质量评估模型为：
    MOS＝5-a*log(max(log(B1),0.01))-b*log(max(log(B2),0.01))-c*log(max(logF,0.01))-d*log(max(log(TI),0.01))；
    所述B1为所述VR视频的码率，所述B2为所述VR视频的分辨率，所述F为所述VR视频的帧率，a,b,c,d为常数。
  13. 一种评估装置，其特征在于，包括：
    存储有可执行程序代码的存储器；
    与所述耦合的处理器；
    所述处理器调用所述存储器中存储的所述可执行程序代码，执行如权利要求1-6任一项所述的方法。
  14. 一种计算机可读存储介质，其特征在于，所述计算机存储介质存储有计算机程序，所述计算机程序包括程序指令，所述程序指令当被处理器执行时使所述处理器执行如权利要求1-6任一项所述的方法。
PCT/CN2020/090724 2019-05-17 2020-05-17 Vr视频质量评估方法及装置 WO2020233536A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
KR1020217039548A KR102600721B1 (ko) 2019-05-17 2020-05-17 Vr 영상 품질 평가 방법 및 장치
EP20810116.2A EP3958558A4 (en) 2019-05-17 2020-05-17 VIRTUAL REALITY VIDEO QUALITY ASSESSMENT METHOD AND DEVICE
JP2021566118A JP7327838B2 (ja) 2019-05-17 2020-05-17 Ｖｒ映像の品質を査定する方法、装置およびコンピュータプログラム
US17/527,604 US20220078447A1 (en) 2019-05-17 2021-11-16 Method and apparatus for assessing quality of vr video

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910416533.0 2019-05-17
CN201910416533.0A CN111953959A (zh) 2019-05-17 2019-05-17 Vr视频质量评估方法及装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/527,604 Continuation US20220078447A1 (en) 2019-05-17 2021-11-16 Method and apparatus for assessing quality of vr video

Publications (1)

Publication Number Publication Date
WO2020233536A1 true WO2020233536A1 (zh) 2020-11-26

Family

ID=73336233

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/090724 WO2020233536A1 (zh) 2019-05-17 2020-05-17 Vr视频质量评估方法及装置

Country Status (6)

Country Link
US (1) US20220078447A1 (zh)
EP (1) EP3958558A4 (zh)
JP (1) JP7327838B2 (zh)
KR (1) KR102600721B1 (zh)
CN (1) CN111953959A (zh)
WO (1) WO2020233536A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112367524B (zh) * 2020-12-08 2022-08-09 重庆邮电大学 一种全景视频编码方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103780901B (zh) * 2014-01-22 2015-08-19 上海交通大学 基于视频空间和时间信息的视频质量及压缩码率估计方法
CN104883563A (zh) * 2011-04-11 2015-09-02 华为技术有限公司 一种视频数据质量评估方法和装置
CN106547352A (zh) * 2016-10-18 2017-03-29 小派科技（上海）有限责任公司 一种虚拟现实画面的显示方法、头戴显示设备及其系统
CN107483920A (zh) * 2017-08-11 2017-12-15 北京理工大学 一种基于多层级质量因子的全景视频评估方法及系统
GB2560156A (en) * 2017-02-17 2018-09-05 Sony Interactive Entertainment Inc Virtual reality system and method
CN109451303A (zh) * 2018-12-24 2019-03-08 合肥工业大学 一种针对VR视频中用户体验质量QoE的建模方法

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100588271C (zh) 2006-08-08 2010-02-03 安捷伦科技有限公司 基于分组度量和图像度量两者测量视频质量的系统和方法
JP4787303B2 (ja) 2008-09-25 2011-10-05 日本電信電話株式会社 映像品質推定装置、方法、およびプログラム
JP5707461B2 (ja) 2013-09-24 2015-04-30 日本電信電話株式会社 映像品質推定装置、映像品質推定方法及びプログラム
JP6228906B2 (ja) 2014-10-31 2017-11-08 日本電信電話株式会社 映像品質推定装置、方法およびプログラム
CN105430383A (zh) * 2015-12-07 2016-03-23 广东电网有限责任公司珠海供电局 一种视频流媒体业务的体验质量评估方法
US10277914B2 (en) 2016-06-23 2019-04-30 Qualcomm Incorporated Measuring spherical image quality metrics based on user field of view
CN107968941A (zh) * 2016-10-19 2018-04-27 中国电信股份有限公司 用于评估视频用户体验的方法和装置
CN106657980A (zh) * 2016-10-21 2017-05-10 乐视控股（北京）有限公司 全景视频的质量测试方法和装置
CN108881894B (zh) * 2017-05-08 2020-01-17 华为技术有限公司 Vr多媒体的体验质量确定方法及装置
US11037531B2 (en) * 2019-10-24 2021-06-15 Facebook Technologies, Llc Neural reconstruction of sequential frames

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104883563A (zh) * 2011-04-11 2015-09-02 华为技术有限公司 一种视频数据质量评估方法和装置
CN103780901B (zh) * 2014-01-22 2015-08-19 上海交通大学 基于视频空间和时间信息的视频质量及压缩码率估计方法
CN106547352A (zh) * 2016-10-18 2017-03-29 小派科技（上海）有限责任公司 一种虚拟现实画面的显示方法、头戴显示设备及其系统
GB2560156A (en) * 2017-02-17 2018-09-05 Sony Interactive Entertainment Inc Virtual reality system and method
CN107483920A (zh) * 2017-08-11 2017-12-15 北京理工大学 一种基于多层级质量因子的全景视频评估方法及系统
CN109451303A (zh) * 2018-12-24 2019-03-08 合肥工业大学 一种针对VR视频中用户体验质量QoE的建模方法

Also Published As

Publication number Publication date
EP3958558A1 (en) 2022-02-23
KR20220003087A (ko) 2022-01-07
JP2022533928A (ja) 2022-07-27
US20220078447A1 (en) 2022-03-10
KR102600721B1 (ko) 2023-11-09
EP3958558A4 (en) 2022-06-15
JP7327838B2 (ja) 2023-08-16
CN111953959A (zh) 2020-11-17

Similar Documents

Publication Publication Date Title
CN111950653B (zh) 视频处理方法和装置、存储介质及电子设备
US20220261960A1 (en) Super-resolution reconstruction method and related apparatus
Moorthy et al. Visual quality assessment algorithms: what does the future hold?
CN109727303B (zh) 视频展示方法、系统、计算机设备、存储介质和终端
WO2016127883A1 (zh) 一种图像区域检测方法及装置
KR102045575B1 (ko) 스마트 미러 디스플레이 장치
CN112562019A (zh) 图像色彩调整方法及装置、计算机可读介质和电子设备
CN111866483B (zh) 颜色还原方法及装置、计算机可读介质和电子设备
EP4036841A1 (en) Image processing method and apparatus
WO2023134523A1 (zh) 内容自适应视频编码方法、装置、设备和存储介质
CN114511041B (zh) 模型训练方法、图像处理方法、装置、设备和存储介质
KR20170066227A (ko) 복수의 비디오들을 제시하는 사용자 인터페이스를 생성하기 위한 방법
WO2020233536A1 (zh) Vr视频质量评估方法及装置
JP2024511103A (ja) 近似値に基づいて画像又はビデオの品質を評価する方法及び装置、第１のモデルの訓練方法及び装置、電子機器、記憶媒体、並びにコンピュータプログラム
CN112565887B (zh) 一种视频处理方法、装置、终端及存储介质
CN114286172A (zh) 数据处理方法及装置
CN112967193A (zh) 图像校准方法及装置、计算机可读介质和电子设备
CN113014960B (zh) 一种在线制作视频的方法、装置及存储介质
Huang et al. Edge device-based real-time implementation of CycleGAN for the colorization of infrared video
WO2018192518A1 (zh) 数据处理方法、装置及存储介质
CN113949899B (zh) 视频质量评估方法及装置
US10026201B2 (en) Image classifying method and image displaying method
CN114630139A (zh) 直播视频的质量评估方法及其相关设备
CN114092359A (zh) 花屏处理方法、装置和电子设备
US10764578B2 (en) Bit rate optimization system and method

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20810116

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021566118

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2020810116

Country of ref document: EP

Effective date: 20211117

ENP Entry into the national phase

Ref document number: 20217039548

Country of ref document: KR

Kind code of ref document: A