CN107438180B - Depth perception quality evaluation method for 3D video - Google Patents
Depth perception quality evaluation method for 3D video Download PDFInfo
- Publication number
- CN107438180B CN107438180B CN201710751972.8A CN201710751972A CN107438180B CN 107438180 B CN107438180 B CN 107438180B CN 201710751972 A CN201710751972 A CN 201710751972A CN 107438180 B CN107438180 B CN 107438180B
- Authority
- CN
- China
- Prior art keywords
- depth
- image
- distortion
- evaluated
- left view
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000008447 perception Effects 0.000 title claims abstract description 43
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000013441 quality evaluation Methods 0.000 title description 5
- 238000001514 detection method Methods 0.000 claims abstract description 9
- 238000011156 evaluation Methods 0.000 claims abstract description 8
- 230000003247 decreasing effect Effects 0.000 claims description 36
- 230000006870 function Effects 0.000 claims description 10
- 230000007935 neutral effect Effects 0.000 claims description 3
- 238000003646 Spearman's rank correlation coefficient Methods 0.000 description 4
- 238000007689 inspection Methods 0.000 description 4
- 230000000007 visual effect Effects 0.000 description 3
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
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
- 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
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N2013/0074—Stereoscopic image analysis
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Image Analysis (AREA)
- Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
Abstract
The present invention relates to a kind of depth perception quality evaluating methods of 3 D video, corresponding with image to be evaluated with reference to 3-D image by choosing；Key point matching is carried out by the left view of image to be evaluated and with reference to the left view of 3-D image, and calculates the depth distortion of image left view to be evaluated；Key point matching is carried out by the right view of image to be evaluated and with reference to the right view of 3-D image, and calculates the depth distortion of image right view to be evaluated；Critical point detection is carried out to image to be evaluated and with reference to 3-D image respectively, and calculates the binocular depth distortion of image to be evaluated；The depth distortion of left view, the depth distortion of right view and binocular depth distortion are merged, depth quality distortion is obtained；Depth quality distortion is converted into depth perception quality, it is thus possible to which distortion evaluation is carried out to image to be evaluated according to depth perception quality.
Description
Technical field
The present invention relates to the distortion of 3 D video evaluations, more particularly to a kind of depth perception quality evaluation of 3 D video
Method.
Background technique
3 D video can be supplied to viewer's more true 3D vision impression, in recent years, 3 D video increasingly by
To liking for spectators.Since the data volume of 3 D video is huge, part has to reduce to save bandwidth when on-line normalization
Image quality.Depth perception is one of most important feature of 3 D stereo video and the main distinction of itself and two-dimensional video.Its matter
Amount, referred to as depth perception quality or depth quality are to directly affect 3 D video together with picture quality and visual comfort
One of big factor of key three of Quality of experience.3 D video is needed to compress and be handled, and is compressed and introduced distortion in treatment process,
These distortions will directly affect depth perception quality again.
Summary of the invention
Based on this, it is necessary to provide a kind of depth perception quality evaluating method of 3 D video.
A kind of depth perception quality evaluating method of 3 D video, comprising the following steps:
It chooses corresponding with image to be evaluated with reference to 3-D image；
Key point matching is carried out by the left view of image to be evaluated and with reference to the left view of 3-D image, and is calculated to be evaluated
The depth of valence image left view is distorted；
Key point matching is carried out by the right view of image to be evaluated and with reference to the right view of 3-D image, and is calculated to be evaluated
The depth of valence image right view is distorted；
Critical point detection is carried out to image to be evaluated and with reference to 3-D image respectively, and calculates the binocular of image to be evaluated
Depth distortion；
The depth distortion of the left view, the depth distortion of the right view and binocular depth distortion are melted
It closes, obtains depth quality distortion；
The depth quality distortion is converted into depth perception quality, and according to the depth perception quality to described to be evaluated
Valence image carries out distortion evaluation.
The left view by image to be evaluated and the left view with reference to 3-D image carry out in one of the embodiments,
Key point matching, and the step of calculating the depth distortion of left view includes:
Calculate the distortion of left view Depth cue；
Obtain the set of keypoints detected；
The lower decreasing concentration of the left view Depth cue is calculated according to the set of keypoints；
The depth distortion of the left view is judged according to the lower decreasing concentration of the left view Depth cue.
In one of the embodiments, it is described calculate left view Depth cue distortion the step of include:
Using formulaCalculate the left view Depth cue
Distortion, wherein φ (I) be receptive field in image Ι monocular vision information, Ω () be comparison function,For figure to be evaluated
The left view of picture,For the corresponding left view with reference to 3-D image.
It is described in one of the embodiments, to be calculated under the left view Depth cue according to the set of keypoints
The step of decreasing concentration includes:
Using formulaCalculate the left view
The lower decreasing concentration of figure Depth cue, wherein Num () is the number of element in set,For distortion 3-D image left view inspection
The set of characteristic points measured,For the set of characteristic points detected with reference to 3-D image left view.
The right view by image to be evaluated and the right view with reference to 3-D image carry out in one of the embodiments,
Key point matching, and the step of calculating the depth distortion of right view includes:
Calculate the distortion of right view Depth cue；
Obtain the set of keypoints detected；
The lower decreasing concentration of the right view Depth cue is calculated according to the set of keypoints；
The depth distortion of the right view is calculated according to the judgement of the lower decreasing concentration of the right view Depth cue.
In one of the embodiments, it is described calculate right view Depth cue distortion the step of include:
Using formulaCalculate the left view Depth cue
Distortion, wherein φ (I) is image Ι monocular vision information in receptive field, and Ω () is comparison function,For image to be evaluated
Left view,For the corresponding left view with reference to 3-D image.
It is described in one of the embodiments, to be calculated under the right view Depth cue according to the set of keypoints
The step of decreasing concentration includes:
Using formulaCalculate the right view
The lower decreasing concentration of figure Depth cue, wherein Num () is the number of element in set,For distortion 3-D image right view inspection
The set of characteristic points measured,For the set of characteristic points detected with reference to 3-D image right view.
It is described in one of the embodiments, to carry out critical point detection to image to be evaluated and with reference to 3-D image respectively,
And the step of calculating the binocular depth distortion of image to be evaluated, includes:
Calculate the distortion of the binocular depth clue of image to be evaluated；
Obtain the set of binocular ranging point pair；
The lower decreasing concentration of the binocular depth clue is calculated according to the set of the binocular ranging point pair；
The depth distortion of the image to be evaluated is judged according to the lower decreasing concentration of the binocular depth clue.
The step of distortion of the binocular depth clue for calculating image to be evaluated in one of the embodiments, includes:
Using formulaIt is deep to calculate the binocular
Spend the distortion of clue, wherein Θ (IL, IR) it is receptive field neutral body image to (IL, IR) binocular vision information, Ω () be than
Compared with function,For image pair to be evaluated,3-D image pair is referred to be corresponding.
The set according to the binocular ranging point pair calculates the binocular depth line in one of the embodiments,
The step of lower decreasing concentration of rope includes:
Using formulaIt is deep to calculate the binocular
Spend the lower decreasing concentration of clue, wherein Num () is the number of element in set, MDisThe characteristic point detected for distortion 3-D image
Set, MRefFor the set of characteristic points detected with reference to 3-D image.
The depth perception quality evaluating method of above-mentioned 3 D video is corresponding with image to be evaluated with reference to three-dimensional by choosing
Image；Key point matching is carried out by the left view of image to be evaluated and with reference to the left view of 3-D image, and is calculated to be evaluated
The depth of image left view is distorted；Key point is carried out by the right view of image to be evaluated and with reference to the right view of 3-D image
Match, and calculates the depth distortion of image right view to be evaluated；It is carried out respectively to image to be evaluated and with reference to 3-D image crucial
Point detection, and calculate the binocular depth distortion of image to be evaluated；By the distortion of the depth of left view, the depth distortion of right view and
Binocular depth distortion is merged, and depth quality distortion is obtained；Depth quality distortion is converted into depth perception quality, thus energy
It is enough that distortion evaluation is carried out to image to be evaluated according to depth perception quality.
Detailed description of the invention
Fig. 1 is the flow chart of the depth perception quality evaluating method of 3 D video；
Fig. 2 is the schematic diagram of the depth perception quality evaluating method of 3 D video.
Specific embodiment
To facilitate the understanding of the present invention, a more comprehensive description of the invention is given in the following sections with reference to the relevant attached drawings.In attached drawing
Give presently preferred embodiments of the present invention.But the invention can be realized in many different forms, however it is not limited to this paper institute
The embodiment of description.On the contrary, purpose of providing these embodiments is keeps the understanding to the disclosure more thorough
Comprehensively.
Unless otherwise defined, all technical and scientific terms used herein and belong to technical field of the invention
The normally understood meaning of technical staff is identical.Term as used herein in the specification of the present invention is intended merely to description tool
The purpose of the embodiment of body, it is not intended that the limitation present invention.Term as used herein "and/or" includes one or more related
Listed item any and all combinations.
As shown in Figure 1, the flow chart of the depth perception quality evaluating method for 3 D video.
A kind of depth perception quality evaluating method of 3 D video, comprising the following steps:
Step S110 chooses corresponding with image to be evaluated with reference to 3-D image.
Due to being to carry out distortion computation processing to image to be evaluated, therefore, it is necessary to choose corresponding 3-D image work
For references object.
Image fault will lead to loss in detail, so as to cause the reduction of monocular and binocular depth clue, i.e. human eye vision system
The receptive field of system visual information relevant to depth perception reduces.Thus, it the monocular depth to image to be evaluated can lose respectively
The distortion of true and binocular depth is calculated, wherein monocular depth distortion refers to the distortion of left view depth and the distortion of right view depth.
Step S120 carries out key point matching by the left view of image to be evaluated and with reference to the left view of 3-D image, and
Calculate the depth distortion of image left view to be evaluated.
Specifically, step S120 includes:
1. calculating the distortion of left view Depth cue.
2. obtaining the set of keypoints detected.
3. calculating the lower decreasing concentration of the left view Depth cue according to the set of keypoints.
4. judging that the depth of the left view is distorted according to the lower decreasing concentration of the left view Depth cue.
Calculate left view Depth cue distortion the step of include:
Using formulaCalculate the left view Depth cue
Distortion, wherein φ (I) be receptive field in image Ι monocular vision information, Ω () be comparison function,For figure to be evaluated
The left view of picture,For the corresponding left view with reference to 3-D image.
The step of calculating the lower decreasing concentration of the left view Depth cue according to the set of keypoints include:
Using formulaCalculate the left view
The lower decreasing concentration of figure Depth cue, wherein Num () is the number of element in set,For distortion 3-D image left view inspection
The set of characteristic points measured,For the set of characteristic points detected with reference to 3-D image left view.
In the present embodiment, the number of key point, that is, monocular vision information is reduced with the aggravation of distortion level.
Step S130 carries out key point matching by the right view of image to be evaluated and with reference to the right view of 3-D image, and
Calculate the depth distortion of image right view to be evaluated.
Specifically, step S130 includes:
1. calculating the distortion of right view Depth cue.
2. obtaining the set of keypoints detected.
3. calculating the lower decreasing concentration of the right view Depth cue according to the set of keypoints.
4. being distorted according to the depth that the judgement of the lower decreasing concentration of the right view Depth cue calculates the right view.
Calculate right view Depth cue distortion the step of include:
Using formulaCalculate the left view Depth cue
Distortion, wherein φ (I) is image Ι monocular vision information in receptive field, and Ω () is comparison function,For image to be evaluated
Left view,For the corresponding left view with reference to 3-D image.
The step of calculating the lower decreasing concentration of the right view Depth cue according to the set of keypoints include:
Using formulaCalculate the right view
The lower decreasing concentration of figure Depth cue, wherein Num () is the number of element in set,For distortion 3-D image right view inspection
The set of characteristic points measured,For the set of characteristic points detected with reference to 3-D image right view.
In the present embodiment, the number of key point, that is, monocular vision information is reduced with the aggravation of distortion level.
Step S140 carries out critical point detection and matching to image to be evaluated and with reference to 3-D image respectively, and calculates
The binocular depth of image to be evaluated is distorted.
Specifically, step S140 includes:
1. calculating the distortion of the binocular depth clue of image to be evaluated.
2. obtaining the set of binocular ranging point pair.
3. calculating the lower decreasing concentration of the binocular depth clue according to the set of the binocular ranging point pair.
4. judging that the depth of the image to be evaluated is distorted according to the lower decreasing concentration of the binocular depth clue.
The step of calculating the distortion of the binocular depth clue of image to be evaluated include:
Using formulaIt is deep to calculate the binocular
Spend the distortion of clue, wherein Θ (IL, IR) it is receptive field neutral body image to (IL, IR) binocular vision information, Ω () be than
Compared with function,For image pair to be evaluated,3-D image pair is referred to be corresponding.
The step of calculating the lower decreasing concentration of the binocular depth clue according to the set of the binocular ranging point pair include:
Using formulaIt is deep to calculate the binocular
Spend the lower decreasing concentration of clue, wherein Num () is the number of element in set, MDisThe characteristic point detected for distortion 3-D image
Set, MRefFor the set of characteristic points detected with reference to 3-D image.
In the present embodiment, the number of key point, that is, monocular vision information is reduced with the aggravation of distortion level.
Step S150 loses the depth distortion of the left view, the depth distortion of the right view and the binocular depth
It is really merged, obtains total depth quality distortion.
In the present embodiment, using formula Δ D=Ψ (Δ DM, Δ DB) calculate total depth quality distortion, wherein Δ DM
Including Δ DMLWith Δ DMR, Ψ () is the fusion function for integrating monocular and binocular depth distortion.
In one embodiment, formula can also be usedCalculate total depth quality distortion, ω1、ω2And ω3To adjust
The weight parameter of whole each element relative importance.
Based on above-mentioned all embodiments, it to be capable of measuring the depth perception distortion of video, key, which is to find, is able to reflect sense
By the monocular of Yezhong stereo-picture and the suitable visual signature of binocular vision information, i.e. φ (I) and Θ (IL, IR).In the present invention
It is middle using the receptive field for using Difference of Gaussian filter analog vision nerve, then with SIFT (Scale-invariant feature
Transform, Scale invariant features transform) description sublist reach.
Since 3 D video is usually shot by horizontally arranged camera, it is therefore necessary to exclude wherein obvious displacement errors
Matching double points.It is vertical that the present embodiment uses random sampling unification algorism (RANdom Sample Consensus, RANSAC) calculating
Affine transformation matrices between body image pair are to exclude abnormal point pair.If the parallax value of certain matching double points has exceeded some
Zone of reasonableness, then these points were corresponding to be excluded before calculating affine transformation matrix.The present embodiment hangs down matching double points
Straight and horizontal parallax threshold value is set to TVAnd TH。
The depth quality distortion is converted into depth perception quality by step S160, and according to the depth perception quality
Distortion evaluation is carried out to the image to be evaluated.
Specifically, depth quality distortion Δ D is converted to depth perception quality DMOS using logistic functionp。
The formula of use are as follows:
In the case where symmetrical distortion and asymmetric distortion, curve matching goodness (R2) it is respectively 0.826 and 0.661.Two
Three-dimensional image the quality evaluation algorithm FI-PSNR and MD3DQA of the current mainstream of kind be used to carry out performance comparison, left and right view
The mean value of the PSNE and SSIM of figure are also used as comparison benchmark.The criterion of evaluation algorithms performance is to calculate the algorithm to predict
Pearson correlation coefficients (PLCC) between DMOSp and practical DMOS, Spearman rank correlation coefficient (SROCC) and root mean square
Error (RMSE).PLCC, SROCC and the RMSE of the present embodiment algorithm on symmetrical distortion data set are respectively 0.909,0.904
With 0.038, asymmetric data and on PLCC, SROCC and RMSE be respectively 0.813,0.750 and 0.036, comparing result is shown in
Table 1.Experiment shows that method disclosed by the invention can accurately pre- depth measurement on symmetrical distortion data set and asymmetric data collection
Spend the decline of perceived quality.
The performance of 1. depth perception quality evaluation algorithm of table compares (optimum of each column is indicated using runic)
Incorporated by reference to Fig. 2.
Based on above-mentioned all embodiments, the specific workflow of the depth perception quality evaluating method of 3 D video is as follows:
Selection is corresponding with image to be evaluated to refer to 3-D image, the left view of left view and image to be evaluated to reference 3-D image
Key point matching is carried out, to calculate the depth distortion of left view.To the right view and image to be evaluated of reference 3-D image
Right view carry out key point matching, thus calculate right view depth distortion.Individually calculate the depth distortion of left and right view
Belong to and calculates monocular depth distortion.And critical point detection directly is carried out to reference 3-D image and image to be evaluated, it will be able to count
Calculate binocular depth distortion.After being merged to monocular depth distortion and binocular depth distortion, image to be evaluated can be carried out
Depth perception quality evaluation.
The depth perception quality evaluating method of above-mentioned 3 D video is corresponding with image to be evaluated with reference to three-dimensional by choosing
Image；Key point matching is carried out by the left view of image to be evaluated and with reference to the left view of 3-D image, and is calculated to be evaluated
The depth of image left view is distorted；Key point is carried out by the right view of image to be evaluated and with reference to the right view of 3-D image
Match, and calculates the depth distortion of image right view to be evaluated；It is carried out respectively to image to be evaluated and with reference to 3-D image crucial
Point detection, and calculate the binocular depth distortion of image to be evaluated；By the distortion of the depth of left view, the depth distortion of right view and
Binocular depth distortion is merged, and depth quality distortion is obtained；Depth quality distortion is converted into depth perception quality, thus energy
It is enough that distortion evaluation is carried out to image to be evaluated according to depth perception quality.
Each technical characteristic of embodiment described above can be combined arbitrarily, for simplicity of description, not to above-mentioned reality
It applies all possible combination of each technical characteristic in example to be all described, as long as however, the combination of these technical characteristics is not deposited
In contradiction, all should be considered as described in this specification.
The embodiments described above only express several embodiments of the present invention, and the description thereof is more specific and detailed, but simultaneously
It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art
It says, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to protection of the invention
Range.Therefore, the scope of protection of the patent of the invention shall be subject to the appended claims.
Claims (10)
1. a kind of depth perception quality evaluating method of 3 D video, which comprises the following steps:
It chooses corresponding with image to be evaluated with reference to 3-D image；
Key point matching is carried out by the left view of image to be evaluated and with reference to the left view of 3-D image, and calculates figure to be evaluated
As the depth of left view is distorted；
Key point matching is carried out by the right view of image to be evaluated and with reference to the right view of 3-D image, and calculates figure to be evaluated
As the depth of right view is distorted；
Critical point detection and matching are carried out to image to be evaluated and with reference to 3-D image respectively, and calculates the double of image to be evaluated
The distortion of mesh depth；
The depth distortion of the left view, the depth distortion of the right view and binocular depth distortion are merged, obtained
Take depth quality distortion；
The depth quality distortion is converted into depth perception quality, and according to the depth perception quality to the figure to be evaluated
As carrying out distortion evaluation.
2. the depth perception quality evaluating method of 3 D video according to claim 1, which is characterized in that it is described will be to be evaluated
What the left view of valence image and the depth for carrying out key point matching with reference to the left view of 3-D image, and calculating left view were distorted
Step includes:
Calculate the distortion of left view Depth cue；
Obtain the set of keypoints detected；
The lower decreasing concentration of the left view Depth cue is calculated according to the set of keypoints；
The depth distortion of the left view is judged according to the lower decreasing concentration of the left view Depth cue.
3. the depth perception quality evaluating method of 3 D video according to claim 2, which is characterized in that the calculating is left
The step of distortion of view Depth cue includes:
Using formulaCalculate the distortion of the left view Depth cue, wherein φ
It (I) is image Ι monocular vision information in receptive field, Ω () is comparison function,For the left view of image to be evaluated,
For the corresponding left view with reference to 3-D image.
4. the depth perception quality evaluating method of 3 D video according to claim 2, which is characterized in that described according to institute
Stating the step of set of keypoints calculates the lower decreasing concentration of the left view Depth cue includes:
Using formulaThe lower decreasing concentration of the left view Depth cue is calculated,
In, Num () is the number of element in set,To be distorted the set of characteristic points that 3-D image left view detects,
For the set of characteristic points detected with reference to 3-D image.
5. the depth perception quality evaluating method of 3 D video according to claim 1, which is characterized in that it is described will be to be evaluated
What the right view of valence image and the depth for carrying out key point matching with reference to the right view of 3-D image, and calculating right view were distorted
Step includes:
Calculate the distortion of right view Depth cue；
Obtain the set of keypoints detected；
The lower decreasing concentration of the right view Depth cue is calculated according to the set of keypoints；
The depth distortion of the right view is calculated according to the judgement of the lower decreasing concentration of the right view Depth cue.
6. the depth perception quality evaluating method of 3 D video according to claim 5, which is characterized in that the calculating is right
The step of distortion of view Depth cue includes:
Using formulaCalculate the distortion of the right view Depth cue, wherein φ
It (I) is image Ι monocular vision information in receptive field, Ω () is comparison function,For the right view of image to be evaluated,
For the corresponding right view with reference to 3-D image.
7. the depth perception quality evaluating method of 3 D video according to claim 5, which is characterized in that described according to institute
Stating the step of set of keypoints calculates the lower decreasing concentration of the right view Depth cue includes:
Using formulaThe lower decreasing concentration of the right view Depth cue is calculated,
In, Num () is the number of element in set,To be distorted the set of characteristic points that 3-D image right view detects,
For the set of characteristic points detected with reference to 3-D image right view.
8. the depth perception quality evaluating method of 3 D video according to claim 1, which is characterized in that described right respectively
Image to be evaluated and reference 3-D image carry out critical point detection and matching, and calculate the binocular depth distortion of image to be evaluated
The step of include:
Calculate the distortion of the binocular depth clue of image to be evaluated；
Obtain the set of binocular ranging point pair；
The lower decreasing concentration of the binocular depth clue is calculated according to the set of the binocular ranging point pair；
The depth distortion of the image to be evaluated is judged according to the lower decreasing concentration of the binocular depth clue.
9. the depth perception quality evaluating method of 3 D video according to claim 8, which is characterized in that it is described calculate to
Evaluate image binocular depth clue distortion the step of include:
Using formulaThe distortion of the binocular depth clue is calculated,
Wherein, Θ (IL, IR) it is receptive field neutral body image to (IL, IR) binocular vision information, Ω () be comparison function,For image pair to be evaluated,3-D image pair is referred to be corresponding.
10. the depth perception quality evaluating method of 3 D video according to claim 8, which is characterized in that the basis
The step of set of the binocular ranging point pair calculates the lower decreasing concentration of the binocular depth clue include:
Using formulaCalculate the lower decreasing concentration of the binocular depth clue, wherein
Num () is the number of element in set, MDisFor the set of characteristic points that distortion 3-D image detects, MRefFor with reference to three-dimensional figure
As the set of characteristic points detected.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710751972.8A CN107438180B (en) | 2017-08-28 | 2017-08-28 | Depth perception quality evaluation method for 3D video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710751972.8A CN107438180B (en) | 2017-08-28 | 2017-08-28 | Depth perception quality evaluation method for 3D video |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107438180A CN107438180A (en) | 2017-12-05 |
| CN107438180B true CN107438180B (en) | 2019-02-22 |
Family
ID=60460710
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710751972.8A Active CN107438180B (en) | 2017-08-28 | 2017-08-28 | Depth perception quality evaluation method for 3D video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107438180B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114648482B (en) * | 2020-12-19 | 2025-03-21 | 中国科学院深圳先进技术研究院 | Quality evaluation method and system for stereoscopic panoramic images |
| CN114841960B (en) * | 2022-05-05 | 2025-04-29 | 咪咕视讯科技有限公司 | Three-dimensional video quality evaluation method, device, equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102170581A (en) * | 2011-05-05 | 2011-08-31 | 天津大学 | Human-visual-system (HVS)-based structural similarity (SSIM) and characteristic matching three-dimensional image quality evaluation method |
| CN102333233A (en) * | 2011-09-23 | 2012-01-25 | 宁波大学 | An Objective Evaluation Method of Stereoscopic Image Quality Based on Visual Perception |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8963998B2 (en) * | 2011-04-15 | 2015-02-24 | Tektronix, Inc. | Full reference system for predicting subjective quality of three-dimensional video |
-
2017
- 2017-08-28 CN CN201710751972.8A patent/CN107438180B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102170581A (en) * | 2011-05-05 | 2011-08-31 | 天津大学 | Human-visual-system (HVS)-based structural similarity (SSIM) and characteristic matching three-dimensional image quality evaluation method |
| CN102333233A (en) * | 2011-09-23 | 2012-01-25 | 宁波大学 | An Objective Evaluation Method of Stereoscopic Image Quality Based on Visual Perception |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107438180A (en) | 2017-12-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105959684B (en) | Stereo image quality evaluation method based on binocular fusion | |
| CN103763552B (en) | Stereoscopic image non-reference quality evaluation method based on visual perception characteristics | |
| CN105744256B (en) | Based on the significant objective evaluation method for quality of stereo images of collection of illustrative plates vision | |
| CN109345502B (en) | Stereo image quality evaluation method based on disparity map stereo structure information extraction | |
| CN102665086B (en) | Method for obtaining parallax by using region-based local stereo matching | |
| CN110909693A (en) | 3D face living body detection method and device, computer equipment and storage medium | |
| CN107578403A (en) | Stereo image quality assessment method based on gradient information to guide binocular view fusion | |
| CN108257165A (en) | Image solid matching method, binocular vision equipment | |
| CN103873854B (en) | The defining method of a kind of stereo-picture subjective assessment subject's quantity and experimental data | |
| CN108596975A (en) | A kind of Stereo Matching Algorithm for weak texture region | |
| CN103780895B (en) | A Stereoscopic Video Quality Evaluation Method | |
| CN103856775B (en) | Processing method for subjective evaluation result of stereo video quality | |
| CN105654142A (en) | Natural scene statistics-based non-reference stereo image quality evaluation method | |
| CN107371016A (en) | No-reference 3D Stereo Image Quality Evaluation Method Based on Asymmetric Distortion | |
| CN107438180B (en) | Depth perception quality evaluation method for 3D video | |
| CN103618891B (en) | Objective evaluation method of stereo camera microspur convergence shooting quality | |
| JP6359985B2 (en) | Depth estimation model generation device and depth estimation device | |
| CN104394405B (en) | A kind of method for evaluating objective quality based on full reference picture | |
| Messai et al. | Deep learning and cyclopean view for no-reference stereoscopic image quality assessment | |
| CN105138979A (en) | Method for detecting the head of moving human body based on stereo visual sense | |
| Wang et al. | A simple quality assessment index for stereoscopic images based on 3d gradient magnitude | |
| Md et al. | Multiscale-ssim index based stereoscopic image quality assessment | |
| CN105631887B (en) | Two-step Parallax Improvement Method and System Based on Adaptive Support Weight Matching Algorithm | |
| Li et al. | Full-reference quality assessment of stereoscopic images using disparity-gradient-phase similarity | |
| CN105869172B (en) | A kind of depth map reliability evaluation Measurement Method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |