CN107371016A - Based on asymmetric distortion without with reference to 3D stereo image quality evaluation methods - Google Patents
Based on asymmetric distortion without with reference to 3D stereo image quality evaluation methods Download PDFInfo
- Publication number
- CN107371016A CN107371016A CN201710613185.7A CN201710613185A CN107371016A CN 107371016 A CN107371016 A CN 107371016A CN 201710613185 A CN201710613185 A CN 201710613185A CN 107371016 A CN107371016 A CN 107371016A
- Authority
- CN
- China
- Prior art keywords
- gabor
- right view
- stereo
- left view
- 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.)
- Pending
Links
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
- H04N2013/0074—Stereoscopic image analysis
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Image Processing (AREA)
Abstract
The present invention relates to it is a kind of based on asymmetric distortion without refer to 3D stereo image quality evaluation methods, comprise the following steps：Using Gabor filter component solution stereo pairs, left view and right view Gabor pyramids are respectively obtained.The one-eyed figure being added after being merged is put pixel-by-pixel with right view Gabor pyramids by left view.Setting value figure is subtracted each other by left view and right view Gabor pyramid absolute differences.One-eyed figure and differential chart are fitted with generalized gaussian model GGD, and extracts statistical nature parameter after fitting, includes average, variance and the form parameter of GGD distributions.The structural similarity SSIM fractions between left view and right view after Gabor filter is decomposed are calculated, in this, as the asymmetrical information of stereo-picture, the statistical nature parameter extracted with reference to previous step, with support vector regression SVR come prognostic chart picture mass fraction.
Description
Technical field：
The present invention relates to carry out evaluating objective quality field to 3D stereo-pictures.
Background technology：
With science and technology make rapid progress develop, the update of science and technology, digital times product progressed into people
Life.Under the tide of digital Age, 3D stereo-pictures field is obtaining fast-developing and application, flat compared to two dimension
Face image, stereo-picture can make people immersively experience third dimension and telepresenc, little by little as multimedia research
Main flow direction.However, often there are various forms of mistakes in stereo-picture during the processing such as encoded, compression, transmission
Very, these distortions can cause the decline of visual quality.Therefore, how picture quality is accurately and effectively evaluated has turned into
The study hotspot of image processing field.
3D stereo-pictures utilize the binocular parallax principle of human eye, and the image from Same Scene is each received by eyes, lead to
Cross brain to merge to form parallax, so as to give user's third dimension and the sense of reality.In the process, binocular vision system serves weight
Act on, because human visual system has extremely complex physiological mechanism, be related to acquiring physics of eyes etc. from psychology
Multi-door cross discipline, thus establish meet human visual system stereo image quality evaluation model be research difficult point.Tradition
On, stereo image quality evaluation (Stereoscopic Image Quality Assessment, SIQA) is broadly divided into two kinds of sides
Method：A kind of method is that left and right view is individually evaluated respectively, and then weighting obtains objective value；Another method is then right
On the basis of the view evaluation of left and right, add parallax or depth information carries out Quality evaluation.The letter of first method implementation
Single, algorithm complex is relatively low, easily realizes；And second method considers parallax information, realize effect and human visual system compared with
To be consistent.
Although current 3D stereo image qualities evaluation method and obtaining bigger progress, most of SIQA algorithms
It is true to have ignored some：1) some SIQA algorithms individually consider the statistical nature and spatial frequency domain feature of natural scene image；
2) existing SIQA methods are more suitable for symmetrical stereo-picture distortion, rather than the effect that symmetrical distortion obtains is unsatisfactory.Therefore,
In the case of asymmetric distortion, the present invention proposes brand-new without referring to 3D stereo image quality evaluation methods.
The content of the invention：
Present invention solves the technical problem that：For asymmetric 3D rendering distortion, propose a kind of new without referring to 3D stereograms
Image quality evaluation method, design Gabor filter carry out pyramid decomposition to stereo-picture or so view, according to retinal rivalry and
Eyes merge principle, and left images are merged to obtain one-eyed figure and difference diagram, then obtained two images are done at normalization
Reason, generalized gaussian model fitting is then carried out, extracts statistical nature respectively, finally carry out quality with support vector regression SVR
Score on Prediction, experimental result show that proposed SIQA algorithms obtain extraordinary effect.Technical solution of the present invention is as follows：
It is a kind of based on asymmetric distortion without refer to 3D stereo image quality evaluation methods, comprise the following steps：
1) Gabor filter component solution stereo pairs are utilized, respectively obtain left view and right view Gabor pyramids.
2) the one-eyed figure being added after being merged is put pixel-by-pixel with right view Gabor pyramids by left view.
3) setting value figure is subtracted each other by left view and right view Gabor pyramid absolute differences.
4) one-eyed figure and differential chart are fitted with generalized gaussian model GGD, and extracts statistical nature parameter after fitting, bag
Include average, variance and the form parameter of GGD distributions.
5) the structural similarity SSIM fractions between left view and right view after Gabor filter is decomposed are calculated, with this
As the asymmetrical information of stereo-picture, the statistical nature parameter extracted with reference to previous step, predicted with support vector regression SVR
Image quality score.
Brief description of the drawings：
By accompanying drawing, the implementation steps and advantage that can make the present invention more highlight, and understand the present invention's with being also easier to
Flow and operation.
The asymmetric distortion 3D rendering left views of Fig. 1 (a), the asymmetric distortion 3D rendering right views of Fig. 1 (b)；
Second layer left view after Fig. 2 (a) Gabor filters are decomposed, the second layer right side regards after Fig. 2 (b) Gabor filters are decomposed
Figure；
Fig. 3 merges one-eyed figure；
Fig. 4 setting value figures；
Fig. 5 normalizes differential chart GGD fitted figures；
The differential chart GGD fitted figures of Fig. 6 difference type of distortion.
Embodiment：
It is convenient to carry out to make the solution of the present invention of greater clarity, in order to more highlight advantages of the present invention and mesh
, embodiment of the present invention is further elaborated and illustrated below in conjunction with the accompanying drawings.
101：Gabor pyramid decompositions
Gabor filter can accurately describe the spatial frequency domain information and local correlations of image.Moreover, Gabor is filtered
Device respond with human eye vision cortical stimulation response and its similar, it is therefore, of the invention to be decomposed using Multiscale Gabor Filters device group
Image.
Firstly, for a given width stereo-picture, we select a width in LIVE 3D rendering databases phase-II white
Noise stereo-picture, it is non-symmetrical distortion, and left and right view distortion level is inconsistent.It is non-symmetrical distortion as shown in Fig. 1 (a)
3D rendering left view, Fig. 1 (b) is non-symmetrical distortion 3D rendering right view.Its left view and right view are used into Gabor respectively
Wave filter is decomposed, and Gabor pyramids is obtained, shown in the expression formula such as formula (1) of Gabor filter：
X'=x cos θ+y sin θs；
Y'=-x sin θ+y cos θ
Wherein x, y represent the space coordinates of pixel in each view, and x', y' are the coordinates after rotation, and λ is wavelength, and θ is
Direction, σ are the parameters on bandwidth, and γ is Gabor filter form parameter, and Ψ is phase offset.
By setting, different filter bandwidhts and direction can produce different scale, the Gabor of different directions is filtered in theory
Ripple device group, but in order to reduce computation complexity, it is horizontal direction that we, which select the direction of Gabor filter, in experiment, is set
As variable parameter, bandwidth is respectively set to { 1,1.64,2.68 } bandwidth.So each view passes through Multiscale Gabor Filters
Result after device component solution just constitutes the Gabor pyramids being made up of three layers of passage, including left and right view.Final figure (1)
By Gabor pyramid decompositions, obtain as shown in Fig. 2 Fig. 2 (a) is the left view of the second layer after Gabor filter decomposition, Fig. 2
(b) right view of the second layer after being decomposed for Gabor filter.Because first layer passage, third layer passage and second layer passage knot
Seemingly, for not burdensome description, we only depict second layer Gabor pyramid decomposition images to fruit.
102：Merge one-eyed figure
In three-dimensional view, retinal rivalry and eyes fusion show important attribute in human-eye visual characteristic, effective 3D
Stereo image quality evaluation is all specifically contemplated that the two factors.Therefore, we are come using one-eyed figure (" Cyclopean " map)
Simulate eyes fusion and the retinal rivalry of human visual system, left view and right view after also Gabor filter is decomposed
Merged, it is exactly one-eyed figure to obtain piece image, and its fusion process is exactly by the left and right view after Gabor pyramid decompositions
Point is added pixel-by-pixel.It is made up of after being decomposed due to Gabor filter 3 layers of passage, so the one-eyed figure after fusion just there are 3 width, i.e.,
Obtain merging one-eyed figure
WhereinThe one-eyed figure being made up of 3 layers of passage is represented respectively.As exist as shown in Figure 3
On second layer passage, merge one-eyed figure and merged to obtain by Fig. 2 (a) left views and Fig. 2 (b) right views.
103：Setting value figure
Differential chart is obtained by the pyramidal left and right view computation differences of Gabor, in order to obtain absolute difference figure, is needed first
The point quadratic sum pixel-by-pixel of the global energy of every layer of Gabor pyramids, i.e. image is calculated, by taking second layer passage as an example, is calculated
It is not E that its left and right view global energy, which is,LAnd ER.Work as EL＞ ERWhen, differential chartIt is exactly the 2nd layer of passage of Gabor pyramids
Left view image vegetarian refreshments subtracts corresponding right view image vegetarian refreshments, works as EL＜ ERWhen, differential chartIt is exactly the 2nd layer of Gabor pyramids
Passage right view image vegetarian refreshments subtracts corresponding left view image vegetarian refreshments.Briefly, differential chart is exactly the absolute difference of left and right view.
Similarly, the differential chart of first passage and third channel is also calculated respectivelyWithSuch as Fig. 4 institutes
Show, be exactly by the differential chart after Fig. 2 (a) and Fig. 2 (b) second layers or so view generation.
104:DM-GGD feature extractions
In order to effectively evaluate the mass fraction of distortion stereo pairs, IQA algorithms need extraction image preferable
Statistical nature.And generalized Gaussian distribution (Generalized Gaussian Distribution, GGD) be one kind can be fine
To describe the prior model of image statisticses feature.Therefore, we are fitted (DM-GGD) with GGD to one-eyed figure and differential chart,
Therefrom Selecting All Parameters extraction characteristics of image.
First, it would be desirable to by one-eyed figureAnd differential chartIt is normalized, calculation formula such as formula
(2), shown in (3)：
Wherein I'C(x,y,fn), I'D(x,y,fn) respectively be normalization after one-eyed figure and differential chart, constant C=0.01
For keeping stability, μ (x, y, fn) and ρ (x, y, fn) be respectively image average and variance, calculation formula such as formula (4), (5)
It is shown：
Ik,l(x,y,fn) what is represented is one-eyed figureOr differential chartThe Gauss weighting function w of circular symmetry
={ wk,l| k=-3 ... 3, l=-3 ... 3 }.
Then, we are used for being fitted I' with GGD modelsC(x,y,fn) or I'D(x,y,fn) coefficient histogram, GGD fitting
Shown in expression formula such as formula (6)：
Above formula parameter (μ, σ2, γ) and it is average, variance and form parameter that GGD is distributed respectively.Wherein,The Γ (1/ γ) of a=β γ/2,
Black column as shown in Figure 5 shows the differential chart after normalized, and red curve is then represented after normalization
The differential chart of GGD fittings, Fig. 6 then describe the differential chart distribution of five kinds of different type of distortion in LIVE databases.It can see
It is different to go out different types of distortion GGD matched curves, while meets 0 average Gaussian Profile.In addition, the degree of bias in statistical nature
(Skewness) calculating formula is s=E (x- μ)3/σ3, kurtosis (Kurtosis) calculating formula k=E (x- μ)4/σ4.Therefore, it is final I
Select GGD be fitted after 4 parameter (σ2, γ, s, k) represent 3D stereo-picture statistical natures, at the same time, one-eyed figure and
Differential chart pyramid haves three layers passage respectively, so the characteristic vector sum that we extract is tieed up for 24 (3 × (4+4)).
104：SVR prognostic chart picture mass fractions
We are using the support vector regression SVR in MATLABL environment LIBSVM tool boxes come prognostic chart picture quality.SVR
It is mainly used to solve optimization problem, shown in its expression-form such as formula (7)：
zj,z'jThe j=1 of >=0, ξ, μ ＞ 0 ... m
Wherein, xjIt is characteristic vector, yjIt is the DMOS values corresponding to training sample, w is weight, φ (xj) it is mapping function,
Parameter zjWith z'jIt is slack variable, Ω is bias, and ξ and η are relevant with training sample.In training, we use RBF
(RBF) the distance between 2 samples of high-dimensional vector space are measured as kernel function.
After extracting feature for each 3D stereo-picture, a characteristic vector is just obtained.We are in LIVE 3D rendering numbers
According to progress SVR regression forecasting experiments on the phase-II of storehouse.LIVE 3D databases phase-II are by 8 with reference to 5 kinds of mistakes of stereo-picture
Proper class type generates 360 pictures, including：White noise, JPEG compression, JP2K compressions, Gaussian Blur and fast-fading.In experiment,
We randomly choose the stereotome of database 80%
1000 times, the median of test data is taken as final quality evaluation fraction.
105：Experimental result
In order to assess the next performance without reference 3D stereo image qualities evaluation proposed by the invention, we select this Pierre
3 graceful rank order correlation coefficient (SROCC), Pearson's linearly dependent coefficient (PLCC) and root-mean-square error (RMSE) evaluations refer to
Mark carrys out evaluation algorithms performance.PLCC, SROCC, RMSE evaluation of estimate such as institute of table 1 corresponding to DM-GGD algorithms proposed by the invention
Show.Wherein, PLCC, SROCC value are bigger represents more consistent with subjective assessment value DMOS, and algorithm effect is better, and RMSE value is then got over
It is small better.As can be seen from the figure DM-GGD algorithms PLCC values can reach 0.920 respectively, and this evaluation result is very high,
Preferable with human eye visual angle system conformance, under independent type of distortion, their expression effect is also fine.
In summary, the present invention mainly just asymmetric 3D stereo-pictures distortion, propose a kind of new without with reference to 3D stereograms
Image quality evaluation method DM-GGD.By experimental result, we have demonstrated that the evaluation method proposed can be good at embodying it is non-
The asymmetrical information of symmetrical 3D stereo-pictures, its performance, which exceedes the overwhelming majority, to be had with reference to and without with reference to SIQA algorithms.Look forward to the future,
Further work from now on is to establish the model for more conforming to human visual system, extracts more accurate characteristics of image to evaluate
Stereo image quality.
Evaluation index PLCC, SROCC, RMSE result of the DM-GGD algorithms of table 1
Claims (1)
1. it is a kind of based on asymmetric distortion without refer to 3D stereo image quality evaluation methods, comprise the following steps：
1) Gabor filter component solution stereo pairs are utilized, respectively obtain left view and right view Gabor pyramids.
2) the one-eyed figure being added after being merged is put pixel-by-pixel with right view Gabor pyramids by left view.
3) setting value figure is subtracted each other by left view and right view Gabor pyramid absolute differences.
4) one-eyed figure and differential chart are fitted with generalized gaussian model GGD, and extracts statistical nature parameter after fitting, including
Average, variance and the form parameter of GGD distributions.
5) the structural similarity SSIM fractions between left view and right view after Gabor filter is decomposed are calculated, in this, as
The asymmetrical information of stereo-picture, the statistical nature parameter extracted with reference to previous step, with support vector regression SVR come prognostic chart picture
Mass fraction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710613185.7A CN107371016A (en) | 2017-07-25 | 2017-07-25 | Based on asymmetric distortion without with reference to 3D stereo image quality evaluation methods |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710613185.7A CN107371016A (en) | 2017-07-25 | 2017-07-25 | Based on asymmetric distortion without with reference to 3D stereo image quality evaluation methods |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107371016A true CN107371016A (en) | 2017-11-21 |
Family
ID=60308628
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710613185.7A Pending CN107371016A (en) | 2017-07-25 | 2017-07-25 | Based on asymmetric distortion without with reference to 3D stereo image quality evaluation methods |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107371016A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108156451A (en) * | 2017-12-11 | 2018-06-12 | 江苏东大金智信息***有限公司 | A kind of 3-D view/video without reference mass appraisal procedure |
| CN108460752A (en) * | 2017-12-20 | 2018-08-28 | 宁波大学 | A kind of asymmetric more distortion objective evaluation method for quality of stereo images |
| CN108648186A (en) * | 2018-05-11 | 2018-10-12 | 北京理工大学 | Based on primary vision perception mechanism without with reference to stereo image quality evaluation method |
| CN109345552A (en) * | 2018-09-20 | 2019-02-15 | 天津大学 | Stereo image quality evaluation method based on region weight |
| CN110634126A (en) * | 2019-04-04 | 2019-12-31 | 天津大学 | No-reference 3D stereo image quality evaluation method based on wavelet packet decomposition |
| CN112233089A (en) * | 2020-10-14 | 2021-01-15 | 西安交通大学 | No-reference stereo mixed distortion image quality evaluation method |
| CN112950629A (en) * | 2021-04-02 | 2021-06-11 | 上海大学 | No-reference panoramic image quality evaluation method and system |
| CN113362315A (en) * | 2021-06-22 | 2021-09-07 | 中国科学技术大学 | Image quality evaluation method and evaluation model based on multi-algorithm fusion |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011109218A (en) * | 2009-11-13 | 2011-06-02 | Victor Co Of Japan Ltd | Stereoscopic video imaging device, and stereoscopic video imaging method |
| CN102722888A (en) * | 2012-05-22 | 2012-10-10 | 天津大学 | Stereoscopic image objective quality evaluation method based on physiological and psychological stereoscopic vision |
| CN105407349A (en) * | 2015-11-30 | 2016-03-16 | 宁波大学 | No-reference objective three-dimensional image quality evaluation method based on binocular visual perception |
| CN105654142A (en) * | 2016-01-06 | 2016-06-08 | 上海大学 | Natural scene statistics-based non-reference stereo image quality evaluation method |
| CN106651835A (en) * | 2016-10-20 | 2017-05-10 | 天津大学 | Entropy-based double-viewpoint reference-free objective stereo-image quality evaluation method |
-
2017
- 2017-07-25 CN CN201710613185.7A patent/CN107371016A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011109218A (en) * | 2009-11-13 | 2011-06-02 | Victor Co Of Japan Ltd | Stereoscopic video imaging device, and stereoscopic video imaging method |
| CN102722888A (en) * | 2012-05-22 | 2012-10-10 | 天津大学 | Stereoscopic image objective quality evaluation method based on physiological and psychological stereoscopic vision |
| CN105407349A (en) * | 2015-11-30 | 2016-03-16 | 宁波大学 | No-reference objective three-dimensional image quality evaluation method based on binocular visual perception |
| CN105654142A (en) * | 2016-01-06 | 2016-06-08 | 上海大学 | Natural scene statistics-based non-reference stereo image quality evaluation method |
| CN106651835A (en) * | 2016-10-20 | 2017-05-10 | 天津大学 | Entropy-based double-viewpoint reference-free objective stereo-image quality evaluation method |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108156451A (en) * | 2017-12-11 | 2018-06-12 | 江苏东大金智信息***有限公司 | A kind of 3-D view/video without reference mass appraisal procedure |
| CN108460752A (en) * | 2017-12-20 | 2018-08-28 | 宁波大学 | A kind of asymmetric more distortion objective evaluation method for quality of stereo images |
| CN108460752B (en) * | 2017-12-20 | 2020-04-10 | 宁波大学 | Objective evaluation method for quality of asymmetric multi-distortion stereo image |
| CN108648186A (en) * | 2018-05-11 | 2018-10-12 | 北京理工大学 | Based on primary vision perception mechanism without with reference to stereo image quality evaluation method |
| CN108648186B (en) * | 2018-05-11 | 2021-11-19 | 北京理工大学 | No-reference stereo image quality evaluation method based on primary visual perception mechanism |
| CN109345552A (en) * | 2018-09-20 | 2019-02-15 | 天津大学 | Stereo image quality evaluation method based on region weight |
| CN110634126A (en) * | 2019-04-04 | 2019-12-31 | 天津大学 | No-reference 3D stereo image quality evaluation method based on wavelet packet decomposition |
| CN112233089A (en) * | 2020-10-14 | 2021-01-15 | 西安交通大学 | No-reference stereo mixed distortion image quality evaluation method |
| CN112950629A (en) * | 2021-04-02 | 2021-06-11 | 上海大学 | No-reference panoramic image quality evaluation method and system |
| CN113362315A (en) * | 2021-06-22 | 2021-09-07 | 中国科学技术大学 | Image quality evaluation method and evaluation model based on multi-algorithm fusion |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107371016A (en) | Based on asymmetric distortion without with reference to 3D stereo image quality evaluation methods | |
| Shao et al. | Full-reference quality assessment of stereoscopic images by learning binocular receptive field properties | |
| CN105744256B (en) | Based on the significant objective evaluation method for quality of stereo images of collection of illustrative plates vision | |
| Song et al. | Real-time exemplar-based face sketch synthesis | |
| CN106462771A (en) | 3D image significance detection method | |
| CN109360178B (en) | Fusion image-based non-reference stereo image quality evaluation method | |
| Fan et al. | Two-layer Gaussian process regression with example selection for image dehazing | |
| CN108334816A (en) | The Pose-varied face recognition method of network is fought based on profile symmetry constraint production | |
| CN112132023A (en) | Crowd counting method based on multi-scale context enhanced network | |
| CN109255358B (en) | 3D image quality evaluation method based on visual saliency and depth map | |
| CN108765414B (en) | No-reference stereo image quality evaluation method based on wavelet decomposition and natural scene statistics | |
| Yue et al. | Blind stereoscopic 3D image quality assessment via analysis of naturalness, structure, and binocular asymmetry | |
| CN104811693B (en) | A kind of stereo image vision comfort level method for objectively evaluating | |
| CN110246111B (en) | No-reference stereoscopic image quality evaluation method based on fusion image and enhanced image | |
| CN104867138A (en) | Principal component analysis (PCA) and genetic algorithm (GA)-extreme learning machine (ELM)-based three-dimensional image quality objective evaluation method | |
| CN108389189B (en) | Three-dimensional image quality evaluation method based on dictionary learning | |
| CN108520510B (en) | No-reference stereo image quality evaluation method based on overall and local analysis | |
| CN111402311A (en) | Knowledge distillation-based lightweight stereo parallax estimation method | |
| CN104866864A (en) | Extreme learning machine for three-dimensional image quality objective evaluation | |
| Karimi et al. | Blind stereo quality assessment based on learned features from binocular combined images | |
| Yang et al. | No-reference stereoimage quality assessment for multimedia analysis towards Internet-of-Things | |
| CN111915589A (en) | Stereo image quality evaluation method based on hole convolution | |
| CN103839244B (en) | Real-time image fusion method and device | |
| CN109523590B (en) | 3D image depth information visual comfort evaluation method based on sample | |
| Bourbia et al. | A multi-task convolutional neural network for blind stereoscopic image quality assessment using naturalness analysis |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171121 |