WO2022044252A1 - Image quality estimation device, image quality estimation method, and program - Google Patents

Image quality estimation device, image quality estimation method, and program Download PDF

Info

Publication number
WO2022044252A1
WO2022044252A1 PCT/JP2020/032587 JP2020032587W WO2022044252A1 WO 2022044252 A1 WO2022044252 A1 WO 2022044252A1 JP 2020032587 W JP2020032587 W JP 2020032587W WO 2022044252 A1 WO2022044252 A1 WO 2022044252A1
Authority
WO
WIPO (PCT)
Prior art keywords
quality
video
tile
image
estimation
Prior art date
Application number
PCT/JP2020/032587
Other languages
French (fr)
Japanese (ja)
Inventor
正憲 小池
和久 山岸
勇一朗 浦田
則次 恵木
Original Assignee
日本電信電話株式会社
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 日本電信電話株式会社 filed Critical 日本電信電話株式会社
Priority to PCT/JP2020/032587 priority Critical patent/WO2022044252A1/en
Publication of WO2022044252A1 publication Critical patent/WO2022044252A1/en

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]
    • H04N21/20—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
    • H04N21/24—Monitoring of processes or resources, e.g. monitoring of server load, available bandwidth, upstream requests

Definitions

  • the present invention relates to a video quality estimation device, a video quality estimation method, and a program.
  • HMDs head-mounted displays
  • VR video distribution services have increased, and opportunities to view virtual reality (VR) videos that can be seen 360 ° using HMDs, smartphones, stationary displays, etc. have increased. ing.
  • VR virtual reality
  • the user views the VR image
  • the user can change the line-of-sight direction by wearing an HMD and performing actions such as shaking the head or moving the body.
  • the viewing direction of the image can be changed by operating the conventional stationary display with a mouse or the like.
  • the quality that a user feels about a video is a parameter (bit rate, resolution, frame rate, quantization parameter (QP)) related to the video quality. ), Etc.).
  • QoE Quality of Experience
  • QP quantization parameter
  • Conventional 2D image quality evaluation technology is under study. For example, in Non-Patent Documents 1 to 3, when an evaluator views a two-dimensional image, the image is displayed. An objective quality evaluation technique for estimating video quality from bit streams, packet headers, metadata, etc. of the above is disclosed.
  • the first method is a uniform distribution method in which the entire image is distributed with uniform image quality, as in the conventional 2D image.
  • the second method is a tile-based method in which the video in the viewing direction of the user displayed on the HMD or the like is delivered in high quality, and the bandwidth is suppressed by delivering or not delivering the video not displayed on the other HMD or the like in low image quality. This is a method called delivery.
  • the tiles sent with high image quality (hereinafter referred to as "high-quality tiles") that are viewed when the line-of-sight direction is not changed. Watch both images of tiles sent in low quality (hereinafter referred to as “low quality tiles”) that are temporarily viewed.
  • the low-quality image is viewed during the period until the tile in the new viewing direction is switched from the low-quality tile to the high-quality tile (hereinafter referred to as "switching delay"). Degradation of image quality is perceived by the user.
  • the implementer can set a parameter called a preset for setting the quality of the coding device and the speed of coding at the time of coding, and can set the quality and speed of coding.
  • a preset for setting the quality of the coding device and the speed of coding at the time of coding
  • the quality and speed of coding For example, in an encoder called ffmpeg, ultrafast, superfast, veryfast, faster, fast, medium, slow, slower, very slow, and placebo can be set as presets in order of increasing encoding speed. Since the encoding setting changes depending on the preset setting, the image quality perceived by the user also changes.
  • video quality is affected by parameters related to video quality such as bit rate, resolution, frame rate, and QP, and a method for estimating video quality using these parameters has been proposed. ..
  • the correspondence between the parameters related to the video quality and the video quality differs depending on the preset settings set when the coding is implemented. For example, in speed-focused implementations, reducing reference frames when calculating motion vectors and reducing bitrate redistribution between frames tends to reduce coding efficiency in areas of high motion. be. Therefore, speed-focused implementations are lower than quality achieved at the same bit rate and QP in quality-focused implementations. In the conventional quality estimation using only the QP of the video, there is a problem that the change in quality due to the implementation method of these codings cannot be extracted.
  • the present invention has been made in view of the above points, and an object of the present invention is to improve the estimation accuracy of the perceived quality of a tile-based VR image by a viewer.
  • the video quality estimation device includes an extraction unit that extracts parameters related to video quality from the tile-based VR video, and the video quality that the viewer of the tile-based VR video experiences based on the parameters. It has an estimation unit that calculates an estimated value, and a correction unit that corrects the estimated value in consideration of the influence of the coding setting on the quality.
  • a user wearing a head-mounted display (HMD) can change the line-of-sight direction of a VR image that can be seen 360 ° by actions such as shaking his head or moving his body.
  • the quality of the image that the user experiences when viewing the image in a state where the user of the conventional stationary display or smartphone can change the viewing direction of the image by operating the mouse or the like QoE (Quality of Experience)
  • QoE Quality of Experience
  • bit stream of encoded data of the delivered video as an input, in addition to parameters related to video quality such as bit rate, resolution, frame rate, and quantization parameter (QP), the difference in implementation can be quantitatively captured from the bit stream.
  • QP quantization parameter
  • a method using the frame type for each frame of the video and the amount of bits allocated to each frame can be considered.
  • the video is composed of a plurality of frames, and the types of video frames are I-frame, which is a reference frame, B-frame, which is compressed and expressed by using the difference in video for each frame, and P.
  • I-frame which is a reference frame
  • B-frame which is compressed and expressed by using the difference in video for each frame
  • P the amount of each of the three types of frames in the video, the amount of bits allocated to each frame, and the like, which have three types of frames, can be changed depending on the coding implementation method.
  • FIG. 1 is a diagram showing a graph showing the amount of bits allocated for each frame for different implementation methods related to coding.
  • FIG. 1 shows the amount of bits allocated for each frame when a VR video A is encoded by varying the bit rate with two settings (mounting X and mounting Y) having different mounting methods. Has been done.
  • Implementation X is a quality-oriented coding method compared to implementation Y, and by increasing the number of frames that refer to the motion vectors of B and P frames, the quality of B and P frames is improved with a small amount of bits. can do. Therefore, when the implementation X and the implementation Y having the same bit rate are compared, the amount of bits allocated to the I frame is larger in the implementation X than in the implementation Y, but the amount of bits allocated to the B and P frames is the amount of the implementation Y. There are more than implementation X. In this way, the amount of bits allocated to each frame differs depending on the implementation method.
  • the high-quality tile is the entire VR video (the entire range of the 360-degree VR video. Since it is an image in the range where a part of "whole") is cut out, the method of allocating the bit amount differs depending on the cut out range. Further, since the motion vector and the like at the cut-out boundary are also different from the low-quality tiles that use the entire image, the high-quality tiles and the low-quality tiles have different bit amount allocations. Therefore, in the present embodiment, the amount of allocated bits for each of the high-quality tile and the low-quality tile is taken into consideration and used as a parameter for capturing the influence of the coding implementation method.
  • FIG. 2 is a diagram showing a hardware configuration example of the video quality estimation device 10 according to the embodiment of the present invention.
  • the video quality estimation device 10 of FIG. 2 has a drive device 100, an auxiliary storage device 102, a memory device 103, a CPU 104, an interface device 105, and the like, which are connected to each other by a bus B, respectively.
  • the program that realizes the processing in the video quality estimation device 10 is provided by a recording medium 101 such as a CD-ROM.
  • a recording medium 101 such as a CD-ROM.
  • the program is installed in the auxiliary storage device 102 from the recording medium 101 via the drive device 100.
  • the program does not necessarily have to be installed from the recording medium 101, and may be downloaded from another computer via the network.
  • the auxiliary storage device 102 stores the installed program and also stores necessary files, data, and the like.
  • the memory device 103 reads the program from the auxiliary storage device 102 and stores it when there is an instruction to start the program.
  • the CPU 104 executes the function related to the video quality estimation device 10 according to the program stored in the memory device 103.
  • the interface device 105 is used as an interface for connecting to a network.
  • FIG. 3 is a diagram showing a functional configuration example of the video quality estimation device 10 according to the embodiment of the present invention.
  • the video quality estimation device 10 includes a video quality parameter extraction unit 11, a base video quality estimation unit 12, a video quality correction unit 13, and the like. Each of these parts is realized by a process of causing the CPU 104 to execute one or more programs installed in the video quality estimation device 10.
  • FIG. 4 is a flowchart for explaining an example of a processing procedure executed by the video quality estimation device 10.
  • the video quality parameter extraction unit 11 inputs the tile-based VR distribution video bitstream and distribution setting parameters to be evaluated for the perceived quality.
  • the distribution setting parameter means a video switching delay in the tile-based VR video. The switching delay is the period until the tile in the new viewing direction is switched from the low-quality tile to the high-quality tile when the user changes the viewing direction.
  • the distribution setting parameter does not necessarily have to be given as an input. Further, the input distribution video bitstream may be the entire distribution video (all sections) or a part (partial section).
  • the video quality parameter extraction unit 11 is referred to as a parameter (hereinafter referred to as "video quality parameter") relating to the video quality of each of the high-quality tile and the low-quality tile from the input distribution video bitstream. .) Is extracted.
  • the video quality parameter is the resolution, frame rate, bit rate, QP of each of the high-quality tile and the low-quality tile in the entire video, and at least one of the allocated bits of each of the I, P, and B frames. Means one. That is, the video quality parameter may be all the parameters listed above, or may be some parameters.
  • the base video quality estimation unit 12 inputs the video quality parameter extracted by the video quality parameter extraction unit 11 and the distribution setting parameter (delivery setting switching delay) input to the video quality parameter extraction unit 11.
  • the base video quality estimated value VQ' corresponding to the video quality parameter is output (S103).
  • the base video quality estimated value VQ' refers to a provisional estimated value (the influence on the quality of the video mounting method is not taken into consideration) regarding the video quality.
  • the base video quality estimation unit 12 first calculates the quality VQ H of high-quality tiles using the following quality estimation formula.
  • br H is the average bit rate of the high-quality tile
  • res H is the resolution of the high-quality tile
  • fr is the frame rate
  • v 1 to v 7 are predetermined coefficients.
  • the base image quality estimation unit 12 uses the same X and Y as above, substitutes the quantization parameter QP H of the high-quality tile in place of the br H of the equation 1, and substitutes the quantization parameter QP H of the high-quality tile into the equation 2 below.
  • the quality VQ H of the image quality tile may be calculated.
  • the base image quality estimation unit 12 can calculate the quality VQL of the low image quality tile using the parameters of the low image quality tile.
  • the values of the coefficients v 1 to v 7 may be the same as or different from those of the high image quality tile.
  • the base video quality estimation unit 12 calculates the base video quality estimation value VQ'using the quality VQ H of the high-quality tile and the quality VQ L of the low-quality tile.
  • the quality deteriorates due to the movement of the line of sight, but the length (duration) of viewing the low-quality tile at the time of the movement of the line of sight differs depending on the length of the switching delay.
  • the ratio of high-quality tiles displayed when the line of sight moves differs depending on the ratio of high-quality tiles covering the entire VR image, switching delay and high-quality tiles are displayed in the calculation of the base image quality estimate VQ'.
  • Ratio SL is used.
  • the base video quality estimation value VQ' is calculated using the following quality estimation formula using the quality VQ H of the high-quality tile and the quality VQ L of the low-quality tile.
  • delay is the switching delay
  • res H is the resolution of the high-quality tile
  • res is the resolution of the entire video
  • d 1 to d 4 are coefficients.
  • the video quality correction unit 13 inputs the video quality parameter extracted by the video quality parameter extraction unit 11 and the base video evaluation value VQ'output from the base video quality estimation unit 12, and sets the video coding.
  • the image quality value VQ whose quality is corrected by the image mounting method is output (S104).
  • the video quality correction unit 13 calculates the VQ by the equation 4 using the total bit amount assigned to each of the I, B, and P frames of the high-quality tile or the low-quality tile.
  • bit I , bit B , and bit P are the total amount of bits allocated to each of the I frame, B frame, and P frame, br is the bit rate of the entire video, and e1 to e4 are predetermined constants. Is.
  • the characteristics of the coding implementation method can be taken into consideration, and the image quality based on the characteristics of the coding implementation method can be taken into consideration.
  • the value can be calculated. It should be noted that the above-mentioned numbers 1 to 4 are merely examples, and may be replaced by other mathematical formulas having the same purpose. Further, the coefficients used in the equations 1 to 4 may be coefficients that change depending on the codec, profile setting, and the like.
  • the present embodiment it is possible to improve the estimation accuracy of the perceived quality by the viewer of the tile-based VR video. For example, in the tile-based VR video to be distributed, even when different encoding implementation settings are mixed, it is possible to estimate in advance (before distribution) the quality that the user will experience when viewing the video. Become.
  • the video quality parameter extraction unit 11 is an example of the extraction unit.
  • the base video quality estimation unit 12 is an example of the estimation unit.
  • the video quality correction unit 13 is an example of the correction unit.
  • Video quality estimation device 11 Video quality parameter extraction unit 12
  • Base video quality estimation unit 13 Video quality correction unit 100
  • Drive device 101 Recording medium 102
  • Memory device 104 CPU 105 Interface device B Bus

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

This image quality estimation device comprises: an extraction unit which extracts a parameter pertaining to image quality from a tile-based VR image; an estimation unit which calculates, on the basis of the parameter, an estimation value of the image quality experienced by a viewer of the tile-based VR image; and a correction unit which performs, on the estimation value, correction considering an influence of an encoding setting on the quality, and thus enhances the accuracy of estimating the viewer experience quality of the tile-based VR image.

Description

映像品質推定装置、映像品質推定方法及びプログラムVideo quality estimation device, video quality estimation method and program
　本発明は、映像品質推定装置、映像品質推定方法及びプログラムに関する。 The present invention relates to a video quality estimation device, a video quality estimation method, and a program.
　近年、ヘッドマウントディスプレイ（ＨＭＤ）の発展により、ＶＲ映像配信サービスが増加し、ＨＭＤやスマートフォン、据え置き型ディスプレイ等を用いて３６０°見渡すことのできる仮想現実（ＶＲ）映像を視聴する機会が増加している。 In recent years, with the development of head-mounted displays (HMDs), VR video distribution services have increased, and opportunities to view virtual reality (VR) videos that can be seen 360 ° using HMDs, smartphones, stationary displays, etc. have increased. ing.
　ユーザ（視聴者）がＶＲ映像を視聴するとき、例えば、ＨＭＤを装着して、ユーザが首を振る、体を動かすなどの行動によって視線方向を変えることができる。また、従来の据え置き型のディスプレイをマウス等で操作することによって映像の視聴方向を変えることができる。 When the user (viewer) views the VR image, for example, the user can change the line-of-sight direction by wearing an HMD and performing actions such as shaking the head or moving the body. In addition, the viewing direction of the image can be changed by operating the conventional stationary display with a mouse or the like.
　一般的に、映像配信サービスにおいて、ユーザが映像に対して感じる品質（体感品質（ＱｏＥ（Quality of Experience））は、映像品質に関連するパラメータ（ビットレート、解像度、フレームレート、量子化パラメータ（ＱＰ）等）に強く影響を受ける。従来の２Ｄ映像の品質評価技術については検討が進められており、例えば、非特許文献１～３において、評価者が２次元の映像を視聴した際に、映像のビットストリーム、パケットヘッダ、メタデータなどから映像品質を推定する客観品質評価技術が開示されている。 Generally, in a video distribution service, the quality that a user feels about a video (QoE (Quality of Experience)) is a parameter (bit rate, resolution, frame rate, quantization parameter (QP)) related to the video quality. ), Etc.). Conventional 2D image quality evaluation technology is under study. For example, in Non-Patent Documents 1 to 3, when an evaluator views a two-dimensional image, the image is displayed. An objective quality evaluation technique for estimating video quality from bit streams, packet headers, metadata, etc. of the above is disclosed.
　一方、ＶＲ映像配信については、例えば、非特許文献４で示されたＭＰＥＧ－ＤＡＳＨと呼ばれる配信方式を用いて、サーバ側で複数のビットレートで符号化したファイルを用意し、帯域に応じて、最適なビットレートの映像を配信する技術がある。 On the other hand, for VR video distribution, for example, using a distribution method called MPEG-DASH shown in Non-Patent Document 4, files encoded at a plurality of bit rates are prepared on the server side, and files encoded at a plurality of bit rates are prepared according to the band. There is a technology to deliver video with the optimum bit rate.
　ＶＲ映像の符号化・配信手法には、大きく分けて２種類の手法がある。第１の手法は、従来の２Ｄ映像と同様に、映像全体を一様な画質で配信する、一様型の配信手法である。第２の手法は、ＨＭＤ等に表示されるユーザの視聴方向の映像を高画質で配信し、その他のＨＭＤ等に表示されない映像を低画質で配信、又は配信しないことによって帯域を抑えるというタイルベース配信と呼ばれる手法である。 There are roughly two types of VR video coding / distribution methods. The first method is a uniform distribution method in which the entire image is distributed with uniform image quality, as in the conventional 2D image. The second method is a tile-based method in which the video in the viewing direction of the user displayed on the HMD or the like is delivered in high quality, and the bandwidth is suppressed by delivering or not delivering the video not displayed on the other HMD or the like in low image quality. This is a method called delivery.
　ＶＲ映像の品質劣化として、ぼけや精細度の低下などによる空間的ひずみ、滑らかさの劣化やちらつきなどの時間的ひずみ、及びエッジ部分のノイズなどによる時空間的ひずみ等による劣化が存在する。 As the quality deterioration of VR images, there are spatial distortion due to blurring and deterioration of fineness, temporal distortion such as deterioration of smoothness and flicker, and deterioration due to spatiotemporal distortion due to noise at the edge portion.
　また、タイルベースＶＲ映像配信においては、ユーザは、視線方向を変えない場合に視聴される、高画質で送られるタイル（以下、「高画質タイル」という。）、及び視線方向を変えた場合に一時的に視聴される、低画質で送られるタイル（以下、「低画質タイル」という。）の両方の映像を視聴する。ユーザが視聴方向を変化させた場合において、新しい視聴方向のタイルを低画質タイルから高画質タイルに切り替わるまでの期間（以下、「切り替え遅延」という。）において、低画質の映像が視聴されため、映像品質の劣化がユーザによって知覚される。 Further, in the tile-based VR video distribution, when the user changes the line-of-sight direction, the tiles sent with high image quality (hereinafter referred to as "high-quality tiles") that are viewed when the line-of-sight direction is not changed. Watch both images of tiles sent in low quality (hereinafter referred to as "low quality tiles") that are temporarily viewed. When the user changes the viewing direction, the low-quality image is viewed during the period until the tile in the new viewing direction is switched from the low-quality tile to the high-quality tile (hereinafter referred to as "switching delay"). Degradation of image quality is perceived by the user.
　更に、映像の符号化を行う際には、符号化の際の計算量の制約により、符号化の実装者によって符号化の計算量の割り振りが決定されている。したがって、実装者ごとに実現される符号化後映像の品質が異なる。例えば、実装者は、符号化の際に、プリセットと呼ばれる符号化装置の品質や符号化の速度を設定するパラメータを設定し、符号化の品質や速度を設定することができる。例えば、ｆｆｍｐｅｇと呼ばれるエンコーダでは、プリセットとしてエンコード速度が速い順に、ｕｌｔｒａｆａｓｔ、ｓｕｐｅｒｆａｓｔ、ｖｅｒｙｆａｓｔ、ｆａｓｔｅｒ、ｆａｓｔ、ｍｅｄｉｕｍ、ｓｌｏｗ、ｓｌｏｗｅｒ、ｖｅｒｙｓｌｏｗ、ｐｌａｃｅｂｏを設定できる。プリセットの設定によりエンコード設定が変わるため、ユーザ知覚する映像品質も変化する。 Further, when coding a video, the allocation of the coding calculation amount is determined by the coding implementer due to the limitation of the calculation amount at the time of coding. Therefore, the quality of the coded video realized by each implementer is different. For example, the implementer can set a parameter called a preset for setting the quality of the coding device and the speed of coding at the time of coding, and can set the quality and speed of coding. For example, in an encoder called ffmpeg, ultrafast, superfast, veryfast, faster, fast, medium, slow, slower, very slow, and placebo can be set as presets in order of increasing encoding speed. Since the encoding setting changes depending on the preset setting, the image quality perceived by the user also changes.
　上記したように、映像品質は、ビットレート、解像度、フレームレート、ＱＰなどの映像品質に関連するパラメータに影響を受けるため、これらのパラメータを利用して映像品質を推定する手法が提案されている。 As described above, video quality is affected by parameters related to video quality such as bit rate, resolution, frame rate, and QP, and a method for estimating video quality using these parameters has been proposed. ..
　一方で、タイルベースＶＲ映像配信において、ユーザは、高画質タイル及び低画質タイルの両方の映像を視聴するため、高品質タイル及び低画質タイルそれぞれの画質が総合的な体感品質（ＱｏＥ）に影響を与える。したがって、高画質タイル及び低画質タイルそれぞれの影響を加味した品質の推定が必要である。 On the other hand, in tile-based VR video distribution, users view both high-quality tiles and low-quality tiles, so the image quality of each of the high-quality tiles and low-quality tiles affects the overall quality of experience (QoE). give. Therefore, it is necessary to estimate the quality in consideration of the influences of the high-quality tiles and the low-quality tiles.
　また、符号化の実装の際に設定するプリセットの設定によって前記映像品質に関するパラメータと映像品質との対応関係が異なる。例えば、速度重視の実装では、動きベクトルを計算する際の参照フレームの削減や、ビットレートのフレーム間の再配分の削減を行うことで、動きの激しい部分での符号化効率が低くなる傾向がある。したがって、速度重視の実装は、品質重視の実装における同一のビットレートやＱＰにおいて実現される品質よりも低くなる。従来の映像のＱＰ等のみを用いる品質推定においては、これらの符号化の実装手法による品質の変化を抽出できないという課題がある。 In addition, the correspondence between the parameters related to the video quality and the video quality differs depending on the preset settings set when the coding is implemented. For example, in speed-focused implementations, reducing reference frames when calculating motion vectors and reducing bitrate redistribution between frames tends to reduce coding efficiency in areas of high motion. be. Therefore, speed-focused implementations are lower than quality achieved at the same bit rate and QP in quality-focused implementations. In the conventional quality estimation using only the QP of the video, there is a problem that the change in quality due to the implementation method of these codings cannot be extracted.
　本発明は、上記の点に鑑みてなされたものであって、タイルベースＶＲ映像の視聴者による体感品質の推定精度を向上させることを目的とする。 The present invention has been made in view of the above points, and an object of the present invention is to improve the estimation accuracy of the perceived quality of a tile-based VR image by a viewer.
　そこで上記課題を解決するため、映像品質推定装置は、タイルベースＶＲ映像から映像品質に関するパラメータを抽出する抽出部と、前記パラメータに基づいて、前記タイルベースＶＲ映像の視聴者が体感する映像品質の推定値を算出する推定部と、前記推定値について、符号化設定による品質への影響が考慮された補正を行う補正部と、を有する。 Therefore, in order to solve the above problems, the video quality estimation device includes an extraction unit that extracts parameters related to video quality from the tile-based VR video, and the video quality that the viewer of the tile-based VR video experiences based on the parameters. It has an estimation unit that calculates an estimated value, and a correction unit that corrects the estimated value in consideration of the influence of the coding setting on the quality.
　タイルベースＶＲ映像の視聴者による体感品質の推定精度を向上させることができる。 It is possible to improve the estimation accuracy of the quality of experience by the viewer of the tile-based VR video.
符号化に関する異なる実装手法に対してフレームごとに割り当てられるビット量を示すグラフを示す図である。It is a figure which shows the graph which shows the bit amount allocated for each frame for different implementation methods about coding. 本発明の実施の形態における映像品質推定装置１０のハードウェア構成例を示す図である。It is a figure which shows the hardware configuration example of the image quality estimation apparatus 10 in embodiment of this invention. 本発明の実施の形態における映像品質推定装置１０の機能構成例を示す図である。It is a figure which shows the functional structure example of the image quality estimation apparatus 10 in embodiment of this invention. 映像品質推定装置１０が実行する処理手順の一例を説明するためのフローチャートである。It is a flowchart for demonstrating an example of the processing procedure executed by the image quality estimation apparatus 10.
　以下、図面に基づいて本発明の実施の形態を説明する。本実施の形態は、３６０°見渡すことのできるＶＲ映像を、ヘッドマウントディスプレイ（ＨＭＤ）を装着したユーザ（視聴者）が首を振る、体を動かすなどの行動によって視線方向を変えることのできる状態、又は従来の据え置き型のディスプレイやスマートフォンのユーザがマウス等の操作によって映像の視聴方向を変えることのできる等の状態で視聴した際に、ユーザが体感する映像の品質（ＱｏＥ（Quality of Experience）を示す値（映像品質値）を推定する技術について説明する。 Hereinafter, embodiments of the present invention will be described with reference to the drawings. In this embodiment, a user (viewer) wearing a head-mounted display (HMD) can change the line-of-sight direction of a VR image that can be seen 360 ° by actions such as shaking his head or moving his body. Or, the quality of the image that the user experiences when viewing the image in a state where the user of the conventional stationary display or smartphone can change the viewing direction of the image by operating the mouse or the like (QoE (Quality of Experience)). A technique for estimating a value (video quality value) indicating the above will be described.
　配信映像の符号化データのビットストリームを入力とし、ビットレート、解像度、フレームレート、量子化パラメータ（ＱＰ）などの映像品質に関連するパラメータに加え、実装の違いをビットストリームから定量的にとらえることが可能なパラメータを抽出することで、符号化の実装手法の影響をとらえた配信映像の推定が可能となる。 Taking a bit stream of encoded data of the delivered video as an input, in addition to parameters related to video quality such as bit rate, resolution, frame rate, and quantization parameter (QP), the difference in implementation can be quantitatively captured from the bit stream. By extracting the possible parameters, it is possible to estimate the delivered video that captures the influence of the encoding implementation method.
　例えば、映像のフレームごとのフレーム種別と、各フレームに割り当てられるビット量を利用する手法が考えられる。映像は、複数のフレームから構成されており、映像のフレームの種類は、基準となるフレームであるＩフレーム、フレームごとの映像の差分を利用することによって圧縮されて表現されるＢフレーム、及びＰフレームの３種類がある、映像における３種類の各フレームの量や、各フレームに割り当てられるビット量などは、符号化の実装手法によって変えることができる。 For example, a method using the frame type for each frame of the video and the amount of bits allocated to each frame can be considered. The video is composed of a plurality of frames, and the types of video frames are I-frame, which is a reference frame, B-frame, which is compressed and expressed by using the difference in video for each frame, and P. The amount of each of the three types of frames in the video, the amount of bits allocated to each frame, and the like, which have three types of frames, can be changed depending on the coding implementation method.
　図１は、符号化に関する異なる実装手法に対してフレームごとに割り当てられるビット量を示すグラフを示す図である。図１には、或るＶＲ映像Ａに対して、実装手法の異なる２つの設定（実装Ｘ及び実装Ｙ）でビットレートを変動させて符号化した場合における、フレームごとに割り当てられるビット量が示されている。 FIG. 1 is a diagram showing a graph showing the amount of bits allocated for each frame for different implementation methods related to coding. FIG. 1 shows the amount of bits allocated for each frame when a VR video A is encoded by varying the bit rate with two settings (mounting X and mounting Y) having different mounting methods. Has been done.
　実装Ｘは、実装Ｙと比較して品質重視の符号化方法となっており、Ｂ及びＰフレームの動きベクトルを参照するフレーム数を増やすことで、Ｂ及びＰフレームを少ないビット量で高品質化することができる。そのため、同じビットレートである実装Ｘと実装Ｙを比較した場合に、Ｉフレームに割り当てられるビット量は実装Ｘの方が実装Ｙより多いが、Ｂ、Ｐフレームに割り当てられるビット量は実装Ｙの方が実装Ｘより多い。このように、実装手法によってフレームごとに割り当てられるビット量が異なる。 Implementation X is a quality-oriented coding method compared to implementation Y, and by increasing the number of frames that refer to the motion vectors of B and P frames, the quality of B and P frames is improved with a small amount of bits. can do. Therefore, when the implementation X and the implementation Y having the same bit rate are compared, the amount of bits allocated to the I frame is larger in the implementation X than in the implementation Y, but the amount of bits allocated to the B and P frames is the amount of the implementation Y. There are more than implementation X. In this way, the amount of bits allocated to each frame differs depending on the implementation method.
　また、タイルベースＶＲ映像では、高画質タイル及び低画質タイルそれぞれでの映像のフレーム割り当てが必要となるが、高画質タイルは、ＶＲ映像全体（３６０度のＶＲ映像の全範囲。以下、「映像全体」という。）の一部を切り出した範囲の映像であるため、切り出した範囲ごとにビット量の割り当て方も異なってくる。また、切り出した境界における動きベクトル等も全体の映像を利用する低画質タイルと異なるため、高画質タイルと低画質タイルとは異なったビット量の割り当てを持つ。そこで、本実施の形態では、高画質タイル及び低画質タイルそれぞれの割り当てビット量を考慮して、符号化の実装手法の影響をとらえるパラメータとして活用する。 Further, in the tile-based VR video, it is necessary to allocate frames for the video in each of the high-quality tile and the low-quality tile, but the high-quality tile is the entire VR video (the entire range of the 360-degree VR video. Since it is an image in the range where a part of "whole") is cut out, the method of allocating the bit amount differs depending on the cut out range. Further, since the motion vector and the like at the cut-out boundary are also different from the low-quality tiles that use the entire image, the high-quality tiles and the low-quality tiles have different bit amount allocations. Therefore, in the present embodiment, the amount of allocated bits for each of the high-quality tile and the low-quality tile is taken into consideration and used as a parameter for capturing the influence of the coding implementation method.
　図２は、本発明の実施の形態における映像品質推定装置１０のハードウェア構成例を示す図である。図２の映像品質推定装置１０は、それぞれバスＢで相互に接続されているドライブ装置１００、補助記憶装置１０２、メモリ装置１０３、ＣＰＵ１０４、及びインタフェース装置１０５等を有する。 FIG. 2 is a diagram showing a hardware configuration example of the video quality estimation device 10 according to the embodiment of the present invention. The video quality estimation device 10 of FIG. 2 has a drive device 100, an auxiliary storage device 102, a memory device 103, a CPU 104, an interface device 105, and the like, which are connected to each other by a bus B, respectively.
　映像品質推定装置１０での処理を実現するプログラムは、ＣＤ－ＲＯＭ等の記録媒体１０１によって提供される。プログラムを記憶した記録媒体１０１がドライブ装置１００にセットされると、プログラムが記録媒体１０１からドライブ装置１００を介して補助記憶装置１０２にインストールされる。但し、プログラムのインストールは必ずしも記録媒体１０１より行う必要はなく、ネットワークを介して他のコンピュータよりダウンロードするようにしてもよい。補助記憶装置１０２は、インストールされたプログラムを格納すると共に、必要なファイルやデータ等を格納する。 The program that realizes the processing in the video quality estimation device 10 is provided by a recording medium 101 such as a CD-ROM. When the recording medium 101 storing the program is set in the drive device 100, the program is installed in the auxiliary storage device 102 from the recording medium 101 via the drive device 100. However, the program does not necessarily have to be installed from the recording medium 101, and may be downloaded from another computer via the network. The auxiliary storage device 102 stores the installed program and also stores necessary files, data, and the like.
　メモリ装置１０３は、プログラムの起動指示があった場合に、補助記憶装置１０２からプログラムを読み出して格納する。ＣＰＵ１０４は、メモリ装置１０３に格納されたプログラムに従って映像品質推定装置１０に係る機能を実行する。インタフェース装置１０５は、ネットワークに接続するためのインタフェースとして用いられる。 The memory device 103 reads the program from the auxiliary storage device 102 and stores it when there is an instruction to start the program. The CPU 104 executes the function related to the video quality estimation device 10 according to the program stored in the memory device 103. The interface device 105 is used as an interface for connecting to a network.
　図３は、本発明の実施の形態における映像品質推定装置１０の機能構成例を示す図である。図３において、映像品質推定装置１０は、映像品質パラメータ抽出部１１、ベース映像品質推定部１２及び映像品質補正部１３等を有する。これら各部は、映像品質推定装置１０にインストールされた１以上のプログラムが、ＣＰＵ１０４に実行させる処理により実現される。 FIG. 3 is a diagram showing a functional configuration example of the video quality estimation device 10 according to the embodiment of the present invention. In FIG. 3, the video quality estimation device 10 includes a video quality parameter extraction unit 11, a base video quality estimation unit 12, a video quality correction unit 13, and the like. Each of these parts is realized by a process of causing the CPU 104 to execute one or more programs installed in the video quality estimation device 10.
　以下、映像品質推定装置１０が実行する処理手順について説明する。図４は、映像品質推定装置１０が実行する処理手順の一例を説明するためのフローチャートである。 Hereinafter, the processing procedure executed by the video quality estimation device 10 will be described. FIG. 4 is a flowchart for explaining an example of a processing procedure executed by the video quality estimation device 10.
　ステップＳ１０１において、映像品質パラメータ抽出部１１は、体感品質の評価対象とするタイルベースＶＲ配信映像ビットストリーム及び配信設定パラメータを入力する。配信設定パラメータは、タイルベースＶＲ映像における映像の切り替え遅延を意味する。切り替え遅延とは、ユーザが視聴方向を変化させた場合において、新しい視聴方向のタイルを低画質タイルから高画質タイルに切り替わるまでの期間をいう。なお、配信設定パラメータは必ずしも入力として与えられなくてもよい。また、入力される配信映像ビットストリームは、配信映像の全部（全区間）でもよいし、一部（一部の区間）でもよい。 In step S101, the video quality parameter extraction unit 11 inputs the tile-based VR distribution video bitstream and distribution setting parameters to be evaluated for the perceived quality. The distribution setting parameter means a video switching delay in the tile-based VR video. The switching delay is the period until the tile in the new viewing direction is switched from the low-quality tile to the high-quality tile when the user changes the viewing direction. The distribution setting parameter does not necessarily have to be given as an input. Further, the input distribution video bitstream may be the entire distribution video (all sections) or a part (partial section).
　続いて、映像品質パラメータ抽出部１１は、入力された配信映像ビットストリームから、高画質タイル及び低画質タイルそれぞれの映像の品質に関する（映像の品質を表す）パラメータ（以下、「映像品質パラメータ」という。）を抽出する。ここで、映像品質パラメータは、映像全体のうちの高画質タイル及び低画質タイルそれぞれの解像度、フレームレート、ビットレート、ＱＰ、更に、Ｉ、Ｐ、Ｂフレームのそれぞれの割り当てビット量のうちの少なくとも一つを意味する。すなわち、映像品質パラメータは、上記に列記した全てのパラメータでもよく、一部のパラメータでもよい。 Subsequently, the video quality parameter extraction unit 11 is referred to as a parameter (hereinafter referred to as "video quality parameter") relating to the video quality of each of the high-quality tile and the low-quality tile from the input distribution video bitstream. .) Is extracted. Here, the video quality parameter is the resolution, frame rate, bit rate, QP of each of the high-quality tile and the low-quality tile in the entire video, and at least one of the allocated bits of each of the I, P, and B frames. Means one. That is, the video quality parameter may be all the parameters listed above, or may be some parameters.
　続いて、ベース映像品質推定部１２は、映像品質パラメータ抽出部１１によって抽出された映像品質パラメータと、映像品質パラメータ抽出部１１に入力された配信設定パラメータ（配信設定の切り替え遅延）を入力し、当該映像品質パラメータと、ベース映像評価値ＶＱ'との関係を表したモデル式を利用して、当該映像品質パラメータに応じたベース映像品質推定値ＶＱ'を出力する（Ｓ１０３）。なお、ベース映像品質推定値ＶＱ'は、映像品質についての暫定的な（映像の実装手法による品質への影響が考慮されない）推定値をいう。 Subsequently, the base video quality estimation unit 12 inputs the video quality parameter extracted by the video quality parameter extraction unit 11 and the distribution setting parameter (delivery setting switching delay) input to the video quality parameter extraction unit 11. Using a model formula expressing the relationship between the video quality parameter and the base video evaluation value VQ', the base video quality estimated value VQ' corresponding to the video quality parameter is output (S103). The base video quality estimated value VQ'refers to a provisional estimated value (the influence on the quality of the video mounting method is not taken into consideration) regarding the video quality.
　ベース映像品質推定部１２は、まず、以下の品質推定式を用いて、高画質タイルの品質ＶＱＨを算出する。 The base video quality estimation unit 12 first calculates the quality VQ H of high-quality tiles using the following quality estimation formula.
Figure JPOXMLDOC01-appb-M000001
 　但し、ｂｒＨは、高画質タイルの平均ビットレート、ｒｅｓＨは、高画質タイルの解像度、ｆｒは、フレームレートを表し、ｖ１～ｖ７は予め定めされる係数である。
Figure JPOXMLDOC01-appb-M000001
 However, br H is the average bit rate of the high-quality tile, res H is the resolution of the high-quality tile, fr is the frame rate, and v 1 to v 7 are predetermined coefficients.
　なお、ベース映像品質推定部１２は、上記と同様のＸ及びＹを用いて、数１のｂｒＨの代わりに高画質タイルの量子化パラメータＱＰＨを、以下の数２に代入して、高画質タイルの品質ＶＱＨを算出してもよい。 In addition, the base image quality estimation unit 12 uses the same X and Y as above, substitutes the quantization parameter QP H of the high-quality tile in place of the br H of the equation 1, and substitutes the quantization parameter QP H of the high-quality tile into the equation 2 below. The quality VQ H of the image quality tile may be calculated.
Figure JPOXMLDOC01-appb-M000002
 　ベース映像品質推定部１２は、同様にして、低画質タイルのパラメータを用いて、低画質タイルの品質ＶＱＬを算出することができる。なお、低画質タイルの品質を算出する場合、係数ｖ１～ｖ７の値は、高画質タイルと同じでもよく、異なってもよい。
Figure JPOXMLDOC01-appb-M000002
 Similarly, the base image quality estimation unit 12 can calculate the quality VQL of the low image quality tile using the parameters of the low image quality tile. When calculating the quality of the low image quality tile, the values of the coefficients v 1 to v 7 may be the same as or different from those of the high image quality tile.
　続いて、ベース映像品質推定部１２は、高画質タイルの品質ＶＱＨと低画質タイルの品質ＶＱＬとを利用して、ベース映像品質推定値ＶＱ'を算出する。タイルベースＶＲ映像配信では、視線移動によって品質が劣化するが、切り替え遅延の長さによって視線移動時の低画質タイルを視聴する長さ（期間）が異なる。また、高画質タイルがＶＲ映像全体をカバーする割合によって視線移動時に表示される高画質タイルの割合が異なるため、ベース映像品質推定値ＶＱ'の計算には、切り替え遅延や高画質タイルが表示される割合ＳＬが用いられる。例えば、ベース映像品質推定値ＶＱ'は、高画質タイルの品質ＶＱＨ、低画質タイルの品質ＶＱＬを用いて以下の品質推定式を用いて算出される。 Subsequently, the base video quality estimation unit 12 calculates the base video quality estimation value VQ'using the quality VQ H of the high-quality tile and the quality VQ L of the low-quality tile. In the tile-based VR video distribution, the quality deteriorates due to the movement of the line of sight, but the length (duration) of viewing the low-quality tile at the time of the movement of the line of sight differs depending on the length of the switching delay. In addition, since the ratio of high-quality tiles displayed when the line of sight moves differs depending on the ratio of high-quality tiles covering the entire VR image, switching delay and high-quality tiles are displayed in the calculation of the base image quality estimate VQ'. Ratio SL is used. For example, the base video quality estimation value VQ'is calculated using the following quality estimation formula using the quality VQ H of the high-quality tile and the quality VQ L of the low-quality tile.
Figure JPOXMLDOC01-appb-M000003
 但し、ｄｅｌａｙは、切り替え遅延、ｒｅｓＨは、高画質タイルの解像度、ｒｅｓは、映像全体の解像度、ｄ１～ｄ４は、係数である。
Figure JPOXMLDOC01-appb-M000003
 However, delay is the switching delay, res H is the resolution of the high-quality tile, res is the resolution of the entire video, and d 1 to d 4 are coefficients.
　続いて、映像品質補正部１３は、映像品質パラメータ抽出部１１によって抽出された映像品質パラメータ及びベース映像品質推定部１２から出力されたベース映像評価値ＶＱ'を入力し、映像の符号化設定を加味して品質の補正するモデル式を用いて、映像の実装手法によって品質を補正した映像品質値ＶＱを出力する（Ｓ１０４）。例えば、映像品質補正部１３は、高画質タイル又は低画質タイルのＩ、Ｂ、Ｐフレームのそれぞれに割り当てられたビット量の合計を用いて、数４によりＶＱを算出する。 Subsequently, the video quality correction unit 13 inputs the video quality parameter extracted by the video quality parameter extraction unit 11 and the base video evaluation value VQ'output from the base video quality estimation unit 12, and sets the video coding. Using a model formula that takes into account and corrects the quality, the image quality value VQ whose quality is corrected by the image mounting method is output (S104). For example, the video quality correction unit 13 calculates the VQ by the equation 4 using the total bit amount assigned to each of the I, B, and P frames of the high-quality tile or the low-quality tile.
Figure JPOXMLDOC01-appb-M000004
 但し、ｂｉｔＩ、ｂｉｔＢ、ｂｉｔＰは、Ｉフレーム、Ｂフレーム、Ｐフレームのそれぞれの割り当てビット量の合計、ｂｒは、映像全体のビットレート、ｅ１～ｅ４は、予め定められた定数である。
Figure JPOXMLDOC01-appb-M000004
 However, bit I , bit B , and bit P are the total amount of bits allocated to each of the I frame, B frame, and P frame, br is the bit rate of the entire video, and e1 to e4 are predetermined constants. Is.
　上記の計算において、Ｉ、Ｂ、ＰフレームのＱＰや割り当てビット量が利用されることで、符号化の実装手法の特徴を考慮することができ、符号化の実装手法の特徴を踏まえた映像品質値の算出が可能である。なお、上記した数１～数４は、あくまで一例であり、同趣旨の他の数式によって代替されてもよい。また、数１～数４において利用された係数は、コーデックや、プロファイル設定等によって変化する係数とされてもよい。 By using the QP of I, B, and P frames and the amount of allocated bits in the above calculation, the characteristics of the coding implementation method can be taken into consideration, and the image quality based on the characteristics of the coding implementation method can be taken into consideration. The value can be calculated. It should be noted that the above-mentioned numbers 1 to 4 are merely examples, and may be replaced by other mathematical formulas having the same purpose. Further, the coefficients used in the equations 1 to 4 may be coefficients that change depending on the codec, profile setting, and the like.
　上述したように、本実施の形態によれば、タイルベースＶＲ映像の視聴者による体感品質の推定精度を向上させることができる。例えば、配信されるタイルベースＶＲ映像において、異なる符号化の実装設定が混在する場合であっても、ユーザが視聴した際にユーザが体感する品質を予め（配信前に）推定することが可能になる。 As described above, according to the present embodiment, it is possible to improve the estimation accuracy of the perceived quality by the viewer of the tile-based VR video. For example, in the tile-based VR video to be distributed, even when different encoding implementation settings are mixed, it is possible to estimate in advance (before distribution) the quality that the user will experience when viewing the video. Become.
　なお、本実施の形態において、映像品質パラメータ抽出部１１は、抽出部の一例である。ベース映像品質推定部１２は、推定部の一例である。映像品質補正部１３は、補正部の一例である。 In the present embodiment, the video quality parameter extraction unit 11 is an example of the extraction unit. The base video quality estimation unit 12 is an example of the estimation unit. The video quality correction unit 13 is an example of the correction unit.
　以上、本発明の実施の形態について詳述したが、本発明は斯かる特定の実施形態に限定されるものではなく、請求の範囲に記載された本発明の要旨の範囲内において、種々の変形・変更が可能である。 Although the embodiments of the present invention have been described in detail above, the present invention is not limited to such specific embodiments, and various modifications are made within the scope of the gist of the present invention described in the claims.・ Can be changed.
１０　　　　　映像品質推定装置
１１　　　　　映像品質パラメータ抽出部
１２　　　　　ベース映像品質推定部
１３　　　　　映像品質補正部
１００　　　　ドライブ装置
１０１　　　　記録媒体
１０２　　　　補助記憶装置
１０３　　　　メモリ装置
１０４　　　　ＣＰＵ
１０５　　　　インタフェース装置
Ｂ　　　　　　バス 10 Video quality estimation device 11 Video quality parameter extraction unit 12 Base video quality estimation unit 13 Video quality correction unit 100 Drive device 101 Recording medium 102 Auxiliary storage device 103 Memory device 104 CPU
105 Interface device B Bus

Claims (7)

  1. 　タイルベースＶＲ映像から映像品質に関するパラメータを抽出する抽出部と、
    　前記パラメータに基づいて、前記タイルベースＶＲ映像の視聴者が体感する映像品質の推定値を算出する推定部と、
    　前記推定値について、符号化設定による品質への影響が考慮された補正を行う補正部と、
    を有することを特徴とする映像品質推定装置。     An extractor that extracts parameters related to video quality from tile-based VR video,
    An estimation unit that calculates an estimated value of the image quality experienced by the viewer of the tile-based VR image based on the parameters, and an estimation unit.
    A correction unit that corrects the estimated value in consideration of the influence of the coding setting on the quality, and a correction unit.
    A video quality estimation device characterized by having.
  2. 　前記抽出部は、前記タイルベースＶＲ映像の高画質タイル及び低画質タイルそれぞれの前記パラメータを抽出する、
    ことを特徴とする請求項１記載の映像品質推定装置。     The extraction unit extracts the parameters of the high-quality tiles and the low-quality tiles of the tile-based VR image.
    The video quality estimation device according to claim 1.
  3. 　前記補正部は、前記タイルベースＶＲ映像のフレーム種別ごとのビット量に基づいて、前記補正を行う、
    ことを特徴とする請求項１又は２記載の映像品質推定装置。     The correction unit performs the correction based on the bit amount for each frame type of the tile-based VR image.
    The video quality estimation device according to claim 1 or 2, wherein the image quality is estimated.
  4. 　タイルベースＶＲ映像から映像品質に関するパラメータを抽出する抽出手順と、
    　前記パラメータに基づいて、前記タイルベースＶＲ映像の視聴者が体感する映像品質の推定値を算出する推定手順と、
    　前記推定値について、符号化設定による品質への影響が考慮された補正を行う補正手順と、
    をコンピュータが実行することを特徴とする映像品質推定方法。     Extraction procedure to extract parameters related to video quality from tile-based VR video,
    An estimation procedure for calculating an estimated value of the image quality experienced by the viewer of the tile-based VR image based on the parameters, and an estimation procedure.
    A correction procedure for correcting the estimated value in consideration of the influence of the coding setting on the quality, and a correction procedure.
    A video quality estimation method characterized by a computer performing.
  5. 　前記抽出手順は、前記タイルベースＶＲ映像の高画質タイル及び低画質タイルそれぞれの前記パラメータを抽出する、
    ことを特徴とする請求項４記載の映像品質推定方法。     The extraction procedure extracts the parameters of the high-quality tiles and the low-quality tiles of the tile-based VR image.
    The video quality estimation method according to claim 4, wherein the image quality is estimated.
  6. 　前記補正手順は、前記タイルベースＶＲ映像のフレーム種別ごとのビット量に基づいて、前記補正を行う、
    ことを特徴とする請求項４又は５記載の映像品質推定方法。     In the correction procedure, the correction is performed based on the bit amount for each frame type of the tile-based VR image.
    The video quality estimation method according to claim 4 or 5, wherein the image quality is estimated.
  7. 　請求項４乃至６いずれか一項記載の映像品質推定方法をコンピュータに実行させることを特徴とするプログラム。 A program characterized by causing a computer to execute the video quality estimation method according to any one of claims 4 to 6.
PCT/JP2020/032587 2020-08-28 2020-08-28 Image quality estimation device, image quality estimation method, and program WO2022044252A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2020/032587 WO2022044252A1 (en) 2020-08-28 2020-08-28 Image quality estimation device, image quality estimation method, and program

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2020/032587 WO2022044252A1 (en) 2020-08-28 2020-08-28 Image quality estimation device, image quality estimation method, and program

Publications (1)

Publication Number Publication Date
WO2022044252A1 true WO2022044252A1 (en) 2022-03-03

Family

ID=80352955

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2020/032587 WO2022044252A1 (en) 2020-08-28 2020-08-28 Image quality estimation device, image quality estimation method, and program

Country Status (1)

Country Link
WO (1) WO2022044252A1 (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2017098604A (en) * 2015-11-18 2017-06-01 日本電信電話株式会社 Video quality estimation device, video quality estimation method and program
JP2020141374A (en) * 2019-03-01 2020-09-03 日本電信電話株式会社 Video quality estimation device, video quality estimation method, and program

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2017098604A (en) * 2015-11-18 2017-06-01 日本電信電話株式会社 Video quality estimation device, video quality estimation method and program
JP2020141374A (en) * 2019-03-01 2020-09-03 日本電信電話株式会社 Video quality estimation device, video quality estimation method, and program

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
MASANORI KOIKE, YUICHIRO URATA, KAZUHISA YAMAGISHI: "A study on objective quality estimation model for tile-based VR video streaming services", IEICE TECHNICAL REPORT, CQ, vol. 118, no. 503 (CQ2018-102), 7 March 2019 (2019-03-07), jp, pages 55 - 59, XP009535109 *
VAN DER HOOFT JEROEN; TORRES VEGA MARIA; PETRANGELI STEFANO; WAUTERS TIM; DE TURCK FILIP: "Quality Assessment for Adaptive Virtual Reality Video Streaming: A Probabilistic Approach on the User’s Gaze", 2019 22ND CONFERENCE ON INNOVATION IN CLOUDS, INTERNET AND NETWORKS AND WORKSHOPS (ICIN), IEEE, 19 February 2019 (2019-02-19), pages 19 - 24, XP033536627, DOI: 10.1109/ICIN.2019.8685904 *
YOUSEF HIBA; FEUVRE JEAN LE; VALENZISE GIUSEPPE; HULUSIC VEDAD: "Video Quality Evaluation for Tile-Based Spatial Adaptation", 2018 IEEE 20TH INTERNATIONAL WORKSHOP ON MULTIMEDIA SIGNAL PROCESSING (MMSP), IEEE, 29 August 2018 (2018-08-29), pages 1 - 6, XP033457790, DOI: 10.1109/MMSP.2018.8547126 *

Similar Documents

Publication Publication Date Title
EP2441265B1 (en) Motion based dynamic resolution multiple bit rate video encoding
CN104096362B (en) The Rate Control bit distribution of video flowing is improved based on player's region-of-interest
WO2009116666A1 (en) Method, device, and program for objectively evaluating video quality
JP4796019B2 (en) Video quality estimation apparatus, method, and program
CA2949478A1 (en) Systems and methods for motion-vector-aided video interpolation using real-time smooth video playback speed variation
CN108810427B (en) Viewpoint-based panoramic video content representation method and apparatus
Nguyen et al. A client-based adaptation framework for 360-degree video streaming
Takeuchi et al. Perceptual quality driven adaptive video coding using JND estimation
JP5221459B2 (en) Video quality estimation apparatus, system, method and program
US11671654B2 (en) Video quality estimation apparatus, video quality estimation method and program
JP5706759B2 (en) Image coding apparatus and image coding method
WO2022044252A1 (en) Image quality estimation device, image quality estimation method, and program
Yeh et al. Temporal video transcoding based on frame complexity analysis for mobile video communication
Ashimov et al. Quality assessment of gaming videos compressed via AV1
Chien et al. Quality driven frame rate optimization for rate constrained video encoding
JP2011205253A (en) Device, method, and program for transcoded video quality objective evaluation
JP5523357B2 (en) Video quality estimation apparatus, method and program
Takagi et al. Subjective video quality estimation to determine optimal spatio-temporal resolution
Moon et al. An Uniformalized Quality Encoding in Cloud Transcoding System
JP7351398B2 (en) VR video encoding parameter calculation device, VR video encoding parameter calculation method, and program
JP5237226B2 (en) Video quality estimation apparatus, method, and program
Karthikeyan et al. Perceptual video quality assessment in H. 264 video coding standard using objective modeling
Takagi et al. Optimized spatial and temporal resolution based on subjective quality estimation without encoding
WO2014065844A1 (en) A cloud-based system for flash content streaming
US11729418B1 (en) Content adaptive encoding

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: 20951504

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20951504

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP