CN105447906A - Method for calculating lighting parameters and carrying out relighting rendering based on image and model - Google Patents
Method for calculating lighting parameters and carrying out relighting rendering based on image and model Download PDFInfo
- Publication number
- CN105447906A CN105447906A CN201510771082.4A CN201510771082A CN105447906A CN 105447906 A CN105447906 A CN 105447906A CN 201510771082 A CN201510771082 A CN 201510771082A CN 105447906 A CN105447906 A CN 105447906A
- Authority
- CN
- China
- Prior art keywords
- image
- model
- illumination
- light source
- scene
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T15/00—3D [Three Dimensional] image rendering
- G06T15/50—Lighting effects
- G06T15/506—Illumination models
Abstract
The invention discloses a method for calculating lighting parameters and carrying out relighting rendering based on an image and a model, comprising the following steps: (1) using a 3D point cloud model of a scene to calculate the normal vector of each 3D point; (2) calculating the light source position and direction vector of an image according to the coordinates and the normal vector of each 3D point and the pixel value of the image; (3) by supposing that the lighting model of the image is a Phong model, calculating the parameters of an energy function in the model based on the light source information obtained in the previous step; (4) calculating the shadow and highlight areas of the image; and (5) giving a target RGB image to be rendered, fusing and rendering the scene and lighting information of the original image into the target image, and outputting a final rendering result. By using the method, the relighting process of special-effects production in film and television post-production is simplified, the relighting rendering result can be output quickly, and a user can preliminarily judge whether an input image is applicable to later virtual-real combination. The problem that rework is caused as the lens and post-production inconsistency cannot be found in the existing film and television production process is solved.
Description
Technical field
The present invention relates to a kind of heavy irradiation rendering method, especially relate to a kind of computed image illumination parameter based on single width RGB image and image scene three-dimensional point cloud model and carry out the method that heavily illumination plays up.
Background technology
Based on the achievement in research that the heavy lighting (IBRL) of image is existing a large amount of in graphics and image processing field, but owing to affecting by complicated illumination variation, as the formation of shade, multiple light courcess are interfered mutually, existing research major part be only limitted to known 3D model or to 3D model modeling after more environmentally light carry out heavily illumination.The method of the existing heavily illumination based on image at present from realizing principle is generally divided into three major types, based on bidirectional reflectance function, based on basis function and based on the method for plenoptic function.
Game and film in the illumination of synthetic technology counterweight have particular/special requirement.Early stage film shooting all can place environment ball for recording the illumination condition of surrounding environment at floor usually, and this is one of method of carrying out heavily illumination based on reflective function.Although have many restrictions based on the heavily illumination of reflective function, foundation as reflection model directly affect heavily illumination effect, all illumination conditions etc. can not be imitated, but processing time required for the heavy lighting based on reflective function is relatively less and adopted by the researcher that majority pursues efficient heavy lighting.Bidirectional reflectance distribution function (BRDF) contains the reflective information of illumination in scene, accordingly, the people such as Oskar [Oskar2007] first propose to calculate in radiative transfer technology in expectation to use light source to cut technology, cut and bidirectional reflectance function according to observability precomputation light source, a large amount of calculating is placed on pre-computation step, thus with mutual speed, heavily illumination is carried out to static scene and play up.The heavy lighting in early stage is all realize when unknown object light source, for the situation that target light source can control, then has another kind of processing scheme.Heavily the various image collecting devices that propose of the researchers of illumination exactly for control light source position, towards etc. make light source parameters known, thus simplify heavy illumination algorithm.As the people such as Xuehong [Xuehong2014] fix on outdoor scene target, under different illumination conditions, obtain the image series under Same Scene with same video camera, set up BRDF model extraction Lighting information and carry out heavily illumination.The people such as Tze [Tze2009] then calculate its RBF to the image sequence obtained with same method and carry out heavily illumination, with reference to BRDF and spherical harmonic function under the analysis Phong illumination model that the people such as Mahajan [Mahajan2008] propose and carry out the achievement in research of heavily illumination.
Based on the heavy lighting of basis function mainly for static scene.Because basic image contains the display effect of object or scene under multiple illumination condition, therefore the rendering result that linear combination can draw object or scene under target illumination condition is carried out to the basic image of a series of acquisition.Owing to needing a large amount of basic image, the heavy lighting based on basis function is applied to raising accuracy of face identification more, carries out the existing great amount of images such as aftertreatment or Video processing and the heavily illumination aspect had higher requirements to illumination to three-dimensional reconstruction.Up-to-date research be the people such as Amr [Amr2014] with facial image training set for basic image, spherical harmonics basis function in extraction target facial image is compared with training dataset with after illumination parameter, select the combination image close to target image and weight, avoid the requirement of rebuilding face 3D model and strict illumination condition, reduce in recognition of face by the barely satisfactory identification error brought of illumination.
Plenoptic function have recorded in any direction, wavelength and the light of optional position under the time, contains the data of a brush dimension.Plenoptic function may be used for the complex scene effect under simulation multiple light courcess or any light source condition.But also due to the complicacy that plenoptic function calculates, utilize plenoptic function to carry out the research of heavily illumination relatively less.Guangwei and Yebin [Guangwei2009] is for the image set of the many illumination of multiple views, be mixed with multi views vision technique (MVS) and heavily illumination (IBL) technology based on image, full light figure is used to reconstruct 3D model and paste texture with the illumination pattern that back video camera obtains to object, under new photoenvironment figure (employing the photodetection picture library of debevec in article) is decomposed 31 base illumination, obtain the intensity of reflected light under new illumination finally by synthesis, complete heavily illumination.
Summary of the invention
The object of the invention is to for the deficiencies in the prior art, provide a kind of and calculate illumination parameter based on image and model and carry out the method that heavily illumination plays up.
The object of the invention is to be achieved through the following technical solutions: a kind ofly calculate illumination parameter based on image and model and carry out the method that heavily illumination plays up, the method comprises the steps:
(1) pending RGB image and corresponding image scene three-dimensional point cloud model is read in, the relation under light source direction and between image rgb value according to scene point cloud and normal vector, set up system of linear equations, utilize the direction vector of Least Square Method light source;
(2) suppose that in scene, object is Lambert body, consider illumination model, set up the energy function of Phong illumination model in conjunction with light source direction and normal vector;
(3) utilize optimization method minimization of energy function, tried to achieve the surround lighting A of each object in image scene by the partial derivative of minimization of energy function
iwith diffuse reflection numerical value D
i;
(4) the surround lighting A calculated by step 3
iwith diffuse reflection numerical value D
ithe shadow and highlight region of computed image, and preserve into image as intermediate result;
(5) calculate the direction of illumination of target RGB image, according to the illumination model of hypothesis the three-dimensional model of original image joined in target image and carry out fusion and play up, export the image of final rendering.
Further, in described step 1, when normal vector is unknown, normal vector
solve as follows: due to the three-dimensional point cloud One's name is legion of whole model, for when preventing from calculating, internal memory overflows, a cloud will be divided into multiple part, every part gets contiguous k point (k gets 3000) here and the covariance matrix put of the every part of note
wherein
in every part k point,
for requiring the point of normal vector, PCA decomposition being carried out to covariance matrix and tries to achieve a little
normal vector
Further, in described step 1, light source direction
solve as follows: the model of input is obtained by common pure 3 D visual method for reconstructing, this model have recorded the coordinate that the three-dimensional point cloud of image scene and each three-dimensional point correspond to certain pixel on image, and namely in model, each three-dimensional point can find a pixel to correspond in the picture.Due to the more than object of possibility in image scene, the illumination tensor of each object is different, before asking light source direction, first to Image Segmentation Using, isolates each different object area and is designated as i.Suppose in scene, to only have a light source, select simple illumination model here
wherein I is the rgb value of image pixel, and ρ is the constant value illumination tensor relevant with object,
for light source direction, P is the volume coordinate of three-dimensional point,
be a P normal vector, T represents the matrix transpose operation of vector.To each pixel in region each in image, set up system of linear equations by above-mentioned illumination model, solve this system of equations and then can obtain light source direction
Further, the illumination model selected in described step 2 is:
Wherein I is the rgb value of image pixel.I
afor ambient light color, I
pfor the brightness of surround lighting.K
a, k
dand k
sbe respectively surround lighting, diffuse reflection and specularity factor.S
p∈ [0,1] is shade, H
pfor high backscatter extinction logarithmic ratio.
for light source direction,
for normal vector.Definition energy function
for:
for image pixel,
represent pixel
on rgb value,
for pixel
the normal vector of corresponding model three-dimensional point.Note ρ is the constant value illumination tensor relevant with object, then correspond to the ambient light value A of each region i in image
i=I
aρ, diffuse reflectance D
i=I
pρ.
Further, the shadow and highlight coefficient of described step 4 Scene calculates according to following scattering model:
Shade
Gao Guang
Wherein parametric t
sand t
nthe positive threshold value regulating image light slippery, manual setting.A
iand D
ifor surround lighting and the diffuse reflection numerical value of region i in image,
represent pixel in image
rgb value.
Further, in described step 5.Final playing up still utilizes at first for solving the illumination model of light source direction.When after the light source direction having scene and the light source information such as shade, Gao Guang, according to illumination model, just the object in this scene can be joined in another target scene, only need to obtain the light source information in target scene again, be updated in illumination model to obtain rendering result finally.
The invention has the beneficial effects as follows: the present invention is extraction model and Lighting information from the RGB image of single width, and be fused on target RGB image and play up, realize the basic fusion process of heavily illumination quickly and easily, facilitate the deficiency in Timeliness coverage camera lens in movies-making process, revise then and there or retake.Traditional heavy lighting is all play up existing image or video in post-processed, the problem being not suitable for special effect making existed in image and video can not be found in real time, cause the later stage to retake camera lens, extend whole production of film and TV process, add cost simultaneously.The present invention borrows the illumination parameter in the three-dimensional model computed image of image scene, guaranteeing whole heavy illumination render process fast simultaneously, show the effect after simple heavily illumination fusion as far as possible exactly, for shooting provides reference, improve the efficiency of whole shooting process.
Accompanying drawing explanation
Fig. 1 is the overview flow chart of the inventive method;
Fig. 2 is the original RGB image of input;
Fig. 3 is the scene three-dimensional point cloud model that the image of input is corresponding;
Fig. 4 is the image shadow region solving gained;
Fig. 5 is the image highlight area solving gained.
Embodiment
The core of the inventive method is the RGB image zooming-out scene three-dimensional point cloud model wherein according to input, utilizes these information set up energy function and minimize it, thus tries to achieve illumination parameter, finally original image model of place is rendered in target image.
Below utilize an embodiment to describe the embodiment of idiographic flow, step following (see Fig. 1):
(1) original RGB image (see Fig. 2) of single width and three-dimensional scenic point cloud model (see Fig. 3) corresponding to image is read in, the relation under light source direction and between image rgb value according to scene point cloud and normal vector, set up system of linear equations, utilize the direction vector of Least Square Method light source;
(2) suppose that in scene, object is Lambert body, consider illumination model, set up the energy function of Phong illumination model in conjunction with light source direction and normal vector;
(3) utilize optimization method minimization of energy function, tried to achieve the surround lighting A of each object in image scene by the partial derivative of minimization of energy function
iwith diffuse reflection numerical value D
i;
(4) the surround lighting A calculated by step 3
iwith diffuse reflection numerical value D
ithe shadow region (see Fig. 4) of computed image and highlight area (see Fig. 5), and preserve into image as intermediate result;
(5) calculate the direction of illumination of target RGB image, according to the illumination model of hypothesis the three-dimensional model of original image joined in target image and carry out fusion and play up, export the image of final rendering.
Claims (6)
1. calculate illumination parameter based on image and model and carry out the method that heavily illumination plays up, it is characterized in that, the method comprises the steps:
(1) pending RGB image and corresponding image scene three-dimensional point cloud model is read in, the relation under light source direction and between image rgb value according to scene point cloud and normal vector, set up system of linear equations, utilize the direction vector of Least Square Method light source;
(2) suppose that in scene, object is Lambert body, consider illumination model, set up the energy function of Phong illumination model in conjunction with light source direction and normal vector;
(3) utilize optimization method minimization of energy function, tried to achieve the surround lighting A of each object in image scene by the partial derivative of minimization of energy function
iwith diffuse reflection numerical value D
i;
(4) the surround lighting A calculated by step 3
iwith diffuse reflection numerical value D
ithe shadow and highlight region of computed image, and preserve into image as intermediate result;
(5) calculate the direction of illumination of target RGB image, according to the illumination model of hypothesis the three-dimensional model of original image joined in target image and carry out fusion and play up, export the image of final rendering.
2. according to claim 1ly a kind ofly calculate illumination parameter based on image and model and carry out the method that heavily illumination plays up, it is characterized in that, in described step 1, when normal vector is unknown, normal vector
solve as follows: due to the three-dimensional point cloud One's name is legion of whole model, for when preventing from calculating, internal memory overflows, a cloud will be divided into multiple part, every part gets contiguous k point (k gets 3000) here and the covariance matrix put of the every part of note
wherein
in every part k point,
for requiring the point of normal vector, PCA decomposition being carried out to covariance matrix and tries to achieve a little
normal vector
3. according to claim 1ly a kind ofly calculate illumination parameter based on image and model and carry out the method that heavily illumination plays up, it is characterized in that, in described step 1, light source direction
solve as follows: the model of input is obtained by common pure 3 D visual method for reconstructing, this model have recorded the coordinate that the three-dimensional point cloud of image scene and each three-dimensional point correspond to certain pixel on image, and namely in model, each three-dimensional point can find a pixel to correspond in the picture.Due to the more than object of possibility in image scene, the illumination tensor of each object is different, before asking light source direction, first to Image Segmentation Using, isolates each different object area and is designated as i.Suppose in scene, to only have a light source, select simple illumination model here
wherein I is the rgb value of image pixel, and ρ is the constant value illumination tensor relevant with object,
for light source direction, P is the volume coordinate of three-dimensional point,
be a P normal vector, T represents the matrix transpose operation of vector.To each pixel in region each in image, set up system of linear equations by above-mentioned illumination model, solve this system of equations and then can obtain light source direction
4. according to claim 1ly a kind ofly calculate illumination parameter based on image and model and carry out the method that heavily illumination plays up, it is characterized in that, the illumination model selected in described step 2 is:
Wherein I is the rgb value of image pixel.I
afor ambient light color, I
pfor the brightness of surround lighting.K
a, k
dand k
sbe respectively surround lighting, diffuse reflection and specularity factor.S
p∈ [0,1] is shade, H
pfor high backscatter extinction logarithmic ratio.
for light source direction,
for normal vector.Definition energy function
for:
for image pixel,
represent pixel
on rgb value,
for pixel
the normal vector of corresponding model three-dimensional point.Note ρ is the constant value illumination tensor relevant with object, then correspond to the ambient light value A of each region i in image
i=I
aρ, diffuse reflectance D
i=I
pρ.
5. according to claim 1ly a kind ofly calculate illumination parameter based on image and model and carry out the method that heavily illumination plays up, it is characterized in that, the shadow and highlight coefficient of described step 4 Scene calculates according to following scattering model:
Shade
Gao Guang
Wherein parametric t
sand t
nthe positive threshold value regulating image light slippery, manual setting.A
iand D
ifor surround lighting and the diffuse reflection numerical value of region i in image,
represent pixel in image
rgb value.
6. according to claim 1ly a kind ofly calculate illumination parameter based on image and model and carry out the method that heavily illumination plays up, it is characterized in that, in described step 5.Final playing up still utilizes at first for solving the illumination model of light source direction.When after the light source direction having scene and the light source information such as shade, Gao Guang, according to illumination model, just the object in this scene can be joined in another target scene, only need to obtain the light source information in target scene again, be updated in illumination model to obtain rendering result finally.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510771082.4A CN105447906B (en) | 2015-11-12 | 2015-11-12 | The method that weight illumination render is carried out based on image and model calculating illumination parameter |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510771082.4A CN105447906B (en) | 2015-11-12 | 2015-11-12 | The method that weight illumination render is carried out based on image and model calculating illumination parameter |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105447906A true CN105447906A (en) | 2016-03-30 |
| CN105447906B CN105447906B (en) | 2018-03-13 |
Family
ID=55558037
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510771082.4A Expired - Fee Related CN105447906B (en) | 2015-11-12 | 2015-11-12 | The method that weight illumination render is carried out based on image and model calculating illumination parameter |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105447906B (en) |
Cited By (32)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106023296A (en) * | 2016-05-27 | 2016-10-12 | 华东师范大学 | Fluid scene illumination parameter calculating method |
| CN106204714A (en) * | 2016-08-01 | 2016-12-07 | 华东师范大学 | Video fluid illumination calculation method based on Phong model |
| CN106570928A (en) * | 2016-11-14 | 2017-04-19 | 河海大学 | Image-based re-lighting method |
| CN107424206A (en) * | 2017-04-14 | 2017-12-01 | 苏州蜗牛数字科技股份有限公司 | A kind of interactive approach that the performance of virtual scene shadow is influenceed using actual environment |
| CN107506714A (en) * | 2017-08-16 | 2017-12-22 | 成都品果科技有限公司 | A kind of method of face image relighting |
| CN107909640A (en) * | 2017-11-06 | 2018-04-13 | 清华大学 | Face weight illumination method and device based on deep learning |
| CN107944420A (en) * | 2017-12-07 | 2018-04-20 | 北京旷视科技有限公司 | The photo-irradiation treatment method and apparatus of facial image |
| CN108364292A (en) * | 2018-03-26 | 2018-08-03 | 吉林大学 | A kind of illumination estimation method based on several multi-view images |
| CN108460841A (en) * | 2018-01-23 | 2018-08-28 | 电子科技大学 | A kind of indoor scene light environment method of estimation based on single image |
| CN108509887A (en) * | 2018-03-26 | 2018-09-07 | 深圳超多维科技有限公司 | A kind of acquisition ambient lighting information approach, device and electronic equipment |
| CN108682041A (en) * | 2018-04-11 | 2018-10-19 | 浙江传媒学院 | A method of multiple light courcess rendering is carried out based on the sampling of matrix ranks and deep learning |
| CN108765537A (en) * | 2018-06-04 | 2018-11-06 | 北京旷视科技有限公司 | A kind of processing method of image, device, electronic equipment and computer-readable medium |
| CN109224448A (en) * | 2018-09-25 | 2019-01-18 | 北京天马时空网络技术有限公司 | A kind of method and apparatus of streamer rendering |
| CN109389113A (en) * | 2018-10-29 | 2019-02-26 | 大连恒锐科技股份有限公司 | A kind of multi-function footprint acquisition equipment |
| CN109448098A (en) * | 2018-09-29 | 2019-03-08 | 北京航空航天大学 | A method of virtual scene light source is rebuild based on individual night scene image of building |
| CN109618472A (en) * | 2018-07-16 | 2019-04-12 | 马惠岷 | Lamp light control method and system |
| CN109785423A (en) * | 2018-12-28 | 2019-05-21 | 广州华多网络科技有限公司 | Image light compensation method, device and computer equipment |
| CN110009723A (en) * | 2019-03-25 | 2019-07-12 | 阿里巴巴集团控股有限公司 | The method for reconstructing and device of environment light source |
| CN111063034A (en) * | 2019-12-13 | 2020-04-24 | 四川中绳矩阵技术发展有限公司 | Time domain interaction method |
| CN111147745A (en) * | 2019-12-30 | 2020-05-12 | 维沃移动通信有限公司 | Shooting method, shooting device, electronic equipment and storage medium |
| CN111798384A (en) * | 2020-06-10 | 2020-10-20 | 武汉大学 | Reverse rendering human face image illumination information editing method |
| CN111815750A (en) * | 2020-06-30 | 2020-10-23 | 深圳市商汤科技有限公司 | Method and device for polishing image, electronic equipment and storage medium |
| CN111968216A (en) * | 2020-07-29 | 2020-11-20 | 完美世界(北京)软件科技发展有限公司 | Volume cloud shadow rendering method and device, electronic equipment and storage medium |
| CN112258622A (en) * | 2020-10-26 | 2021-01-22 | 北京字跳网络技术有限公司 | Image processing method, image processing device, readable medium and electronic equipment |
| CN112819941A (en) * | 2021-03-05 | 2021-05-18 | 网易（杭州）网络有限公司 | Method, device, equipment and computer-readable storage medium for rendering water surface |
| CN112819940A (en) * | 2021-01-29 | 2021-05-18 | 网易（杭州）网络有限公司 | Rendering method and device and electronic equipment |
| CN112927342A (en) * | 2021-02-22 | 2021-06-08 | 中铁二院工程集团有限责任公司 | Illumination calculation method and fixed pipeline rendering and programmable pipeline rendering methods |
| WO2021226862A1 (en) * | 2020-05-13 | 2021-11-18 | Shanghaitech University | Neural opacity point cloud |
| CN113920036A (en) * | 2021-12-14 | 2022-01-11 | 武汉大学 | Interactive relighting editing method based on RGB-D image |
| WO2022042470A1 (en) * | 2020-08-31 | 2022-03-03 | 浙江商汤科技开发有限公司 | Image decomposition method and related apparatus and device |
| WO2022140887A1 (en) * | 2020-12-28 | 2022-07-07 | 华为技术有限公司 | Image processing method and apparatus |
| CN116385614A (en) * | 2023-03-29 | 2023-07-04 | 深圳海拓时代科技有限公司 | 3D vision module rendering control system based on visualization |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050080602A1 (en) * | 2003-10-10 | 2005-04-14 | Microsoft Corporation | Systems and methods for all-frequency relighting using spherical harmonics and point light distributions |
| CN101246600A (en) * | 2008-03-03 | 2008-08-20 | 北京航空航天大学 | Method for real-time generating reinforced reality surroundings by spherical surface panoramic camera |
| WO2009143163A2 (en) * | 2008-05-21 | 2009-11-26 | University Of Florida Research Foundation, Inc. | Face relighting from a single image |
| CN103035025A (en) * | 2012-12-28 | 2013-04-10 | 浙江大学 | Material high realistic rendering algorithm based on bidirectional reflectance distribution function (BRDF) measured data |
-
2015
- 2015-11-12 CN CN201510771082.4A patent/CN105447906B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050080602A1 (en) * | 2003-10-10 | 2005-04-14 | Microsoft Corporation | Systems and methods for all-frequency relighting using spherical harmonics and point light distributions |
| CN101246600A (en) * | 2008-03-03 | 2008-08-20 | 北京航空航天大学 | Method for real-time generating reinforced reality surroundings by spherical surface panoramic camera |
| WO2009143163A2 (en) * | 2008-05-21 | 2009-11-26 | University Of Florida Research Foundation, Inc. | Face relighting from a single image |
| CN103035025A (en) * | 2012-12-28 | 2013-04-10 | 浙江大学 | Material high realistic rendering algorithm based on bidirectional reflectance distribution function (BRDF) measured data |
Non-Patent Citations (3)
| Title |
|---|
| GD FINLAYSON ET AL: "Removing shadows from images", 《EUROPEAN CONFERENCE ON COMPUTER VISION》 * |
| 丁晓东: "基于图像的重光照技术", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 王晨昊 等: "光学遥感图像重光照方法研究", 《测绘通报》 * |
Cited By (52)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106023296A (en) * | 2016-05-27 | 2016-10-12 | 华东师范大学 | Fluid scene illumination parameter calculating method |
| CN106023296B (en) * | 2016-05-27 | 2018-09-28 | 华东师范大学 | Fluid scene illumination parameter computational methods |
| CN106204714A (en) * | 2016-08-01 | 2016-12-07 | 华东师范大学 | Video fluid illumination calculation method based on Phong model |
| CN106204714B (en) * | 2016-08-01 | 2019-02-01 | 华东师范大学 | Video fluid illumination calculation method based on Phong model |
| CN106570928A (en) * | 2016-11-14 | 2017-04-19 | 河海大学 | Image-based re-lighting method |
| CN106570928B (en) * | 2016-11-14 | 2019-06-21 | 河海大学 | A kind of heavy illumination method based on image |
| CN107424206A (en) * | 2017-04-14 | 2017-12-01 | 苏州蜗牛数字科技股份有限公司 | A kind of interactive approach that the performance of virtual scene shadow is influenceed using actual environment |
| CN107424206B (en) * | 2017-04-14 | 2020-09-22 | 苏州蜗牛数字科技股份有限公司 | Interaction method for influencing shadow expression of virtual scene by using real environment |
| CN107506714A (en) * | 2017-08-16 | 2017-12-22 | 成都品果科技有限公司 | A kind of method of face image relighting |
| CN107909640A (en) * | 2017-11-06 | 2018-04-13 | 清华大学 | Face weight illumination method and device based on deep learning |
| CN107909640B (en) * | 2017-11-06 | 2020-07-28 | 清华大学 | Face relighting method and device based on deep learning |
| CN107944420B (en) * | 2017-12-07 | 2020-10-27 | 北京旷视科技有限公司 | Illumination processing method and device for face image |
| CN107944420A (en) * | 2017-12-07 | 2018-04-20 | 北京旷视科技有限公司 | The photo-irradiation treatment method and apparatus of facial image |
| CN108460841A (en) * | 2018-01-23 | 2018-08-28 | 电子科技大学 | A kind of indoor scene light environment method of estimation based on single image |
| CN108509887A (en) * | 2018-03-26 | 2018-09-07 | 深圳超多维科技有限公司 | A kind of acquisition ambient lighting information approach, device and electronic equipment |
| CN108364292B (en) * | 2018-03-26 | 2021-05-25 | 吉林大学 | Illumination estimation method based on multiple visual angle images |
| CN108364292A (en) * | 2018-03-26 | 2018-08-03 | 吉林大学 | A kind of illumination estimation method based on several multi-view images |
| CN108682041A (en) * | 2018-04-11 | 2018-10-19 | 浙江传媒学院 | A method of multiple light courcess rendering is carried out based on the sampling of matrix ranks and deep learning |
| CN108682041B (en) * | 2018-04-11 | 2021-12-21 | 浙江传媒学院 | Method for performing multi-light-source rendering based on matrix row and column sampling and deep learning |
| CN108765537A (en) * | 2018-06-04 | 2018-11-06 | 北京旷视科技有限公司 | A kind of processing method of image, device, electronic equipment and computer-readable medium |
| CN109618472A (en) * | 2018-07-16 | 2019-04-12 | 马惠岷 | Lamp light control method and system |
| CN109224448A (en) * | 2018-09-25 | 2019-01-18 | 北京天马时空网络技术有限公司 | A kind of method and apparatus of streamer rendering |
| CN109448098B (en) * | 2018-09-29 | 2023-01-24 | 北京航空航天大学 | Method for reconstructing virtual scene light source based on single night scene image of building |
| CN109448098A (en) * | 2018-09-29 | 2019-03-08 | 北京航空航天大学 | A method of virtual scene light source is rebuild based on individual night scene image of building |
| CN109389113A (en) * | 2018-10-29 | 2019-02-26 | 大连恒锐科技股份有限公司 | A kind of multi-function footprint acquisition equipment |
| CN109389113B (en) * | 2018-10-29 | 2020-12-15 | 大连恒锐科技股份有限公司 | Multifunctional footprint acquisition equipment |
| CN109785423A (en) * | 2018-12-28 | 2019-05-21 | 广州华多网络科技有限公司 | Image light compensation method, device and computer equipment |
| CN109785423B (en) * | 2018-12-28 | 2023-10-03 | 广州方硅信息技术有限公司 | Image light supplementing method and device and computer equipment |
| CN110009723A (en) * | 2019-03-25 | 2019-07-12 | 阿里巴巴集团控股有限公司 | The method for reconstructing and device of environment light source |
| CN110009723B (en) * | 2019-03-25 | 2023-01-31 | 创新先进技术有限公司 | Reconstruction method and device of ambient light source |
| CN111063034B (en) * | 2019-12-13 | 2023-08-04 | 四川中绳矩阵技术发展有限公司 | Time domain interaction method |
| CN111063034A (en) * | 2019-12-13 | 2020-04-24 | 四川中绳矩阵技术发展有限公司 | Time domain interaction method |
| CN111147745B (en) * | 2019-12-30 | 2021-11-30 | 维沃移动通信有限公司 | Shooting method, shooting device, electronic equipment and storage medium |
| CN111147745A (en) * | 2019-12-30 | 2020-05-12 | 维沃移动通信有限公司 | Shooting method, shooting device, electronic equipment and storage medium |
| US11727628B2 (en) | 2020-05-13 | 2023-08-15 | Shanghaitech University | Neural opacity point cloud |
| WO2021226862A1 (en) * | 2020-05-13 | 2021-11-18 | Shanghaitech University | Neural opacity point cloud |
| CN111798384A (en) * | 2020-06-10 | 2020-10-20 | 武汉大学 | Reverse rendering human face image illumination information editing method |
| CN111815750A (en) * | 2020-06-30 | 2020-10-23 | 深圳市商汤科技有限公司 | Method and device for polishing image, electronic equipment and storage medium |
| CN111968216B (en) * | 2020-07-29 | 2024-03-22 | 完美世界(北京)软件科技发展有限公司 | Volume cloud shadow rendering method and device, electronic equipment and storage medium |
| CN111968216A (en) * | 2020-07-29 | 2020-11-20 | 完美世界(北京)软件科技发展有限公司 | Volume cloud shadow rendering method and device, electronic equipment and storage medium |
| WO2022042470A1 (en) * | 2020-08-31 | 2022-03-03 | 浙江商汤科技开发有限公司 | Image decomposition method and related apparatus and device |
| CN112258622A (en) * | 2020-10-26 | 2021-01-22 | 北京字跳网络技术有限公司 | Image processing method, image processing device, readable medium and electronic equipment |
| WO2022140887A1 (en) * | 2020-12-28 | 2022-07-07 | 华为技术有限公司 | Image processing method and apparatus |
| CN112819940A (en) * | 2021-01-29 | 2021-05-18 | 网易（杭州）网络有限公司 | Rendering method and device and electronic equipment |
| CN112819940B (en) * | 2021-01-29 | 2024-02-23 | 网易（杭州）网络有限公司 | Rendering method and device and electronic equipment |
| CN112927342B (en) * | 2021-02-22 | 2022-12-20 | 中铁二院工程集团有限责任公司 | Illumination calculation method and fixed pipeline rendering and programmable pipeline rendering methods |
| CN112927342A (en) * | 2021-02-22 | 2021-06-08 | 中铁二院工程集团有限责任公司 | Illumination calculation method and fixed pipeline rendering and programmable pipeline rendering methods |
| CN112819941A (en) * | 2021-03-05 | 2021-05-18 | 网易（杭州）网络有限公司 | Method, device, equipment and computer-readable storage medium for rendering water surface |
| CN112819941B (en) * | 2021-03-05 | 2023-09-12 | 网易（杭州）网络有限公司 | Method, apparatus, device and computer readable storage medium for rendering water surface |
| CN113920036A (en) * | 2021-12-14 | 2022-01-11 | 武汉大学 | Interactive relighting editing method based on RGB-D image |
| CN116385614A (en) * | 2023-03-29 | 2023-07-04 | 深圳海拓时代科技有限公司 | 3D vision module rendering control system based on visualization |
| CN116385614B (en) * | 2023-03-29 | 2024-03-01 | 深圳海拓时代科技有限公司 | 3D vision module rendering control system based on visualization |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105447906B (en) | 2018-03-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105447906A (en) | Method for calculating lighting parameters and carrying out relighting rendering based on image and model | |
| US10692277B1 (en) | Dynamically estimating lighting parameters for positions within augmented-reality scenes using a neural network | |
| CN107341853B (en) | Virtual-real fusion method and system for super-large virtual scene and dynamic screen shooting | |
| Hall et al. | A testbed for realistic image synthesis | |
| CN102096941B (en) | Consistent lighting method under falsehood-reality fused environment | |
| CN111968215B (en) | Volume light rendering method and device, electronic equipment and storage medium | |
| US8928662B2 (en) | Apparatus, method, and system for demonstrating a lighting solution by image rendering | |
| CN108460841A (en) | A kind of indoor scene light environment method of estimation based on single image | |
| Li et al. | Physically-based editing of indoor scene lighting from a single image | |
| CN104077802A (en) | Method for improving displaying effect of real-time simulation image in virtual scene | |
| CN103995700A (en) | Method for achieving global illumination of 3D game engine | |
| Sheng et al. | A spatially augmented reality sketching interface for architectural daylighting design | |
| Grosch et al. | Consistent interactive augmentation of live camera images with correct near-field illumination | |
| Ignatenko et al. | A Real-Time 3D Rendering System with BRDF Materials and Natural Lighting | |
| CN116894922A (en) | Night vision image generation method based on real-time graphic engine | |
| Thompson et al. | Real-time mixed reality rendering for underwater 360 videos | |
| Happa et al. | Studying illumination and cultural heritage | |
| CN114139249A (en) | Automatic light distribution method and device based on illusion engine and electronic equipment | |
| CN114139250A (en) | Automatic light distribution method, device, equipment and storage medium based on illusion engine | |
| Sheng et al. | Virtual heliodon: Spatially augmented reality for architectural daylighting design | |
| CN112258621A (en) | Method for observing three-dimensional rendering two-dimensional animation in real time | |
| Navvab et al. | Evaluation of historical museum interior lighting system using fully immersive virtual luminous environment | |
| Nasman et al. | Physical avatars in a projector-camera tangible user interface enhance quantitative simulation analysis and engagement | |
| Debelov et al. | Light mesh: soft shadows as interpolation of visibility | |
| Li et al. | Translucent material transfer based on single images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180313 Termination date: 20181112 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |