CN104952103B - The shadow map creation method that a kind of viewpoint relies on - Google Patents
The shadow map creation method that a kind of viewpoint relies on Download PDFInfo
- Publication number
- CN104952103B CN104952103B CN201510255989.5A CN201510255989A CN104952103B CN 104952103 B CN104952103 B CN 104952103B CN 201510255989 A CN201510255989 A CN 201510255989A CN 104952103 B CN104952103 B CN 104952103B
- Authority
- CN
- China
- Prior art keywords
- space
- vertebra
- shadow
- cut
- shade
- 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.)
- Expired - Fee Related
Links
Landscapes
- Image Generation (AREA)
Abstract
The invention discloses the shadow map creation method that a kind of viewpoint relies on, and view is cut into vertebra first and is cut to a shade real space, is then based on this shade real space to create shadow map.It is concretely comprised the following steps：(1), view cuts vertebra and cut：The non-shadow space that view is cut to vertebra part is deleted, and so as to form a tightened up shade encircling space, the space after this is cut is named as shade real space, and it is the minimum space for generating shadow map；(2), shadow map generates：Firstly generate light and cut vertebra, then by splitting above-mentioned shade real space, render the object in light section vertebra using shade camera and obtain echo.By the method for the present invention, non-shadow space is eliminated, solves the problems, such as the space overlap between echo；The resolution factor for ensuring echo and screen pixels by two-way partitioning algorithm matches；Vertebra cutting-out method is cut using view and removes non-shadow space, which greatly enhances the space availability ratio of echo.
Description
Technical field
The invention belongs to computer graphics techniques field, more particularly to a kind of shadow map establishment side of viewpoint dependence
Method.
Background technology
Shade is very important visual signature, and shade gives the information about light source position in scene, highlights thing
Spatial relationship between body.Therefore, Shading Rendering is a hot issue in computer graphics, especially in 3d gaming
It is particularly important with the sense of reality for strengthening virtual scene in animation.These years, many methods have been proposed in Many researchers
To solve the problems, such as real-time rendering.Shadow map algorithm (shadow mapping) is calculated as a kind of effective Shading Rendering
Method has been widely used.Standard shadow textures are the algorithms based on image, so when the lack of resolution of shadow map, tradition
Shadow map algorithm generation shade can cause crenellated phenomena.Traditional shadow map algorithm divides view by thresholding method
Cut vertebra and create shadow map, non-shadow space can occupy a big chunk ratio in shadow map, and this is to cause shadow map
Resolution ratio reduce a reason.The reason for another causes the resolution ratio of shadow map to reduce is between shadow map
Space overlap.
Shadow map algorithm has simplicity and high efficiency, and Stamminger is in 2002 in " Perspective
Shadow maps. ", Martin are 2004 in " Anti-aliasing and continuity with
In trapezoidal shadow maps. ", shadow map, these sides are created by specifying truncated cone to be converted in light source space
The problem of method is present be：Only a shadow map can not match the resolution ratio of a large scene.Fan Zhang,Hanqiu
Sun, Leilei Xu, Lee Kit Lun are in 2006 in " Parallel-split shadow maps for large-
A kind of PSSMs (parallel separation shadow map) algorithm, Rouslan are proposed in scale virtual environments "
Dimitrov is in 2008 " Cascaded Shadow Maps " propose a kind of CSM (lamination shadow map) algorithm, and this two
Kind method belongs to thresholding method, exists due to the problem of space overlap causes shadow quality to reduce between shadow map.
Andrew Lauritzen, Marco Salvi, and Aaron Lefohn are in 2011 in " Sample distribution
A kind of auto-partition algorithm by analyzing light source space depth profile is proposed in shadow maps. ", with first two algorithm
Compare, the algorithm improves the utilization rate that shadow map excludes empty shade, but this method still has asking for space overlap
Topic, it is very limited for the scene full of shadow object of complexity, improvement of this method to quality.In a word, traditional scene relies on
The method for creating shadow map cuts the separation of vertebra, shade sample of these methods in spatial distribution based on thresholding method to view
Aspect has advantage, but is the failure to solve the problems, such as that overlapping caused resolution ratio is low.It is limited to two that improve shadow quality
Aspect：1st, space overlap：It is to be cut the AABB bounding boxs of vertebra by view or surrounded ball to generate that light, which cuts vertebra, and light cuts the sky of vertebra
Between it is overlapping be it is inevitable, because space overlap causes the resolution ratio of shadow map to reduce；2nd, non-shadow space：Non-shadow is empty
Between include horizontal plane under space and horizontal plane on there is no the space for including shade object, these non-shadow spaces are for shade wash with watercolours
It is useless for dye, but they can still occupy a big chunk ratio of shadow map, cause the space of shadow map
Utilization rate seriously reduces.
The content of the invention
It is an object of the invention to provide the shadow map creation method that a kind of viewpoint relies on, to solve non-shadow space proportion
Greatly, the problem of space overlap between shadow map.
To achieve the above object, the technical solution adopted by the present invention is：
The shadow map creation method that a kind of viewpoint relies on, it is a shade visibility region that it is accurate that view is cut into vertebra first,
This shade visibility region is then based on to create shadow map.
It is concretely comprised the following steps：
(1), view cuts vertebra and cut
The non-shadow space that view is cut to vertebra part is deleted, will so as to form the shadow object on a tightened up border
Space after this cuts is named as shade visibility region, and it is the minimum space for generating shadow map；
(2), shadow map generates
Firstly generate light and cut vertebra, then by splitting above-mentioned shade real space, generate shadow view.
Delete the space without projected objects on the space for including being located under horizontal plane and horizontal plane in the non-shadow space.
View field is divided using two-way partitioning algorithm, cuts vertebra according to these partition creating light successively.
The viewport that light cuts vertebra is determined that the viewport of light section vertebra determines the viewport of shadow map again by its respective partition.
Vertebra is cut for light, its remote and hither plane should be by determining including all shadow objects inside its ken.
After light cuts vertebra determination, render the object in light section vertebra using shade camera and generate shadow map.
The beneficial effects of the invention are as follows：The present invention obtains shade real space by cutting view section vertebra, can for shade
See that the projection in ray space in space is split, shadow map is created using segmentation result.By the present invention method,
Non-shadow space is eliminated, solves the problems, such as the space overlap between shadow map；Shade is ensured by two-way partitioning algorithm
Figure matching screen pixels distribution；Vertebra method of cutting out is cut using view and removes non-shadow space, which greatly enhances shadow map
Space availability ratio；Compared with conventional method, the shadow map created by the method for the present invention has higher resolution ratio, shade
Can under any visual angle seamless rendering.Shade shake is can't see in viewpoint moving process.In addition, method of the invention is realized
It is relatively simple, it can apply to other programs.
Brief description of the drawings
Fig. 1 is that view cuts in vertebra the schematic diagram for deleting non-shadow space；
Fig. 2 is to create the schematic diagram that light cuts vertebra on light projection plane based on subregion；
Fig. 3 is the schematic diagram of view field longitudinal direction (left side) and horizontal (right side) subregion；
Fig. 4 is when observer is close to ground surface, and quality versus's figure is presented in (left side) of the invention and CSM (right side) shade；
Fig. 5 is when observer is close to ground surface, and Detail contrast figure is presented in (left side) of the invention and CSM (right side) shade；
Fig. 6 is (left side) the of the invention and contrast of CSM (right side) shade precision and seam when observer is away from ground surface
Figure.
Embodiment
The present invention is further described below in conjunction with the accompanying drawings.
The problem of traditional shadow map algorithm can not reduce space overlap, it is because they create in a manner of scene relies on
Shadow map is built, when creating shadow map, whole view is cut vertebra first and separated by these methods, is then cut by these views
Vertebra segmentation section creates light and cuts vertebra.By conventional method, the viewport of shadow map cuts vertebra by light and determined, the resolution of shadow map
Rate increase is limited to thresholding method.
The present invention proposes the shadow map establishment algorithm that a kind of viewpoint relies on and solves the problems, such as space overlap, to improve shade
Quality, in the method for the invention, view is cut into vertebra first and is cut to a shade visibility region, being then based on this shade can
Region is seen to create shadow map.
The shadow map creation method that the viewpoint of the present invention relies on comprises the following steps：
(1), view cuts vertebra and cut
By view cut vertebra part non-shadow space delete, non-shadow space delete include be located at horizontal plane under space and
Space without projected objects on horizontal plane；
As shown in figure 1, non-shadow space includes the space without projected objects on the space under horizontal plane and horizontal plane, these
Space is useless for Shading Rendering, them should be cut in vertebra from view and cropped, closer so as to form one
Shade bounding box, the space after this is cut is named as shade real space, and it is the minimum for generating shadow map
Space.
(2), shadow map generates
Firstly generate light and cut vertebra, cut vertebra by splitting above-mentioned shade visibility region using light, generate shadow view.
Shade real space after refinement is no longer a common truncated cone, and traditional view cuts vertebra partition method can not
Space for this irregular shape of subregion, it is therefore desirable to which a new method is split to above-mentioned shade visibility region.
Above-mentioned shade real space is split to realize, we have proposed the dividing method that viewpoint relies on,
, it is necessary to introduce the concept of light projection plane to understand the method in the method.
Light projection plane is any one plane vertical with light, and shade real space is projected in the plane,
Ray cast plane forms the border of a view field；Only object is located in the view field of ray space, its depth
Value can just be rendered into shadow map.
As shown in Fig. 2 view field is divided into multiple subregions, view field is divided using two-way partitioning algorithm, successively root
Vertebra is cut according to these partition creating light.Light cut vertebra viewport determined by its respective partition, therefore the viewport of shadow map also by
Its corresponding subregion determines.Vertebra is cut for light, its remote and hither plane should be by including all shadow objects inside its ken
To determine.After light cuts vertebra determination, render the object in light section vertebra using shade camera and generate shadow map.
Light, which cuts vertebra, can be considered as the segmentation of shade real space, and its viewport size is according to their distance phases with viewpoint
Close, avoid the space overlap problem that light is cut between vertebra.
Compared with conventional method, the shadow map created by the method for the present invention has the resolution more matched with screen pixels
Rate.
The size of all shadow maps could be arranged to identical value, in this embodiment, be arranged to 2048 × 2048.
The method of view field's subregion is described in detail below：
By two-way partitioning algorithm, a suitable viewport is created for each shade camera.
As shown in figure 3, by the algorithm, view field is longitudinally split first, then horizontal partition again.
Such as Fig. 3, by view field make it is simple regard as one it is symmetrical trapezoidal, it is necessary to define one before detailed description
A little symbols are used for following discussion.
θ (θ ∈ [0, pi/2)) be that observation is square and the angle on one side of view field, its value with the position of observer and
Change；
N is the length of the near side (ns) of view field, and f is the length of its distal edge；
L is the length of view field；
SiFor the subregion of view field, for a subregion Si(i ＞ 1), its vertically and horizontally side are W respectivelyiAnd Hi；
ρ is the space waste factor of each subregion, it represent outside view field partition space with entirely dividing into
Ratio.
It is longitudinally split：
When view field carries out longitudinally split, the wasting space of each subregion is farthest limited；
ζ represents the maximum space factor for allowing waste in each subregion, and the length of side of each subregions of ρ≤ζ becomes in geometric progression
Change,
In this case, corresponding echo precision will not suddenly change, increment coefficient σ by equation (1) Suo Shi,
Subregion SiThe length on two sides is：
Equation (2) can calculate the length of side of each subregion, and it can easily prove out the waste in each subregion
Space is not more than ζ.
NHIt is longitudinal partition number, it should meet the following requirements：
NHAs observer changes and dynamic change, this ensures that shadow map creates in a dynamic way.To save internal memory,
Rendering efficiency is improved, at utmost limits vertical partition quantity, NH≤NHmax(in the present embodiment, NHmax=4),
For subregionIts length of side is calculated as：
In practice,Size be likely larger than other subregions, its wasting space may exceed ζ.
But becauseRemote observer, only account for the screen space of very little ratio.Even if the resolution ratio of shadow map is not
Foot, will not also have a negative impact to shadow quality.
Horizontal partition：
For shadow map, the difference of its uv resolution ratio should be limited to a certain limit ε,
This limitation is necessary, because can ensure that shade can all have higher resolution factor in uv directions, by vertical
To segmentation, to meet this limitation.
According to longitudinally split scheme, for each region Si, WiNot less than Hi。
In this case, it is only necessary to horizontal partition, as shown in Fig. 3 bottom rights.
Horizontal partition NWCorrespond to Si, calculate
NWBe it is conditional, in the present embodiment, NW≤2.The reasonability of this limitation, which is repeated experiment, to be proved.
Behind horizontal partition, the length of side W of each lower section subregionijIt is calculated as
In order to show advantages of the present invention, contrasted with CSM (cascade shadow map) algorithm.
In present embodiment, maximum space factor ζ=0.3 wasted, resolution ratio uv difference pole are allowed in each subregion
It is 0.5 to limit ε.
CSM view cuts vertebra and is divided into 4 parts, and enables PCF Gaussian blurring functions,
As shown in Figures 4 and 5, the present invention is compared with conventional method, and when observer is close to ground surface, method of the invention is shown
More shadow details.
Such as Fig. 6, when observer is away from ground surface, method of the invention can also keep high shadow quality, and not have
Any shade seam.
The VSM of the present invention can realize the shade non-jitter in observer rocks.
In addition, efficiency of the invention is very high, in our computer accessory Core I7 CPU, 8G memory, and
In the case of the GPU of GTX 780, average frame rate about 90fps.
Shade real space is projected ray cast plane by the present invention first, the algorithm pair then relied on using viewpoint
View field carries out two-way segmentation, and the viewport of shadow map is determined by these subregions；Then, based on these partition creating light
Line cuts vertebra.In this case, the viewport of light section vertebra is determined by projection subregion.It is of the invention compared with conventional method
Method creates shadow map by opposite order：
Projection localization region of the shadow map based on shade real space creates, and solves space overlap and non-shadow space
The problem of；To solve the problems, such as non-shadow space, view is cut into vertebra and is cut to shade real space；In order to solve space overlap
Problem, shade real space is projected into ray cast plane, and the algorithm that viewpoint relies on according to viewpoint is carried out to view field
Split.
Described above is only the preferred embodiment of the present invention, it should be pointed out that：For the ordinary skill people of the art
For member, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications also should
It is considered as protection scope of the present invention.
Claims (7)
- A kind of 1. shadow map creation method that viewpoint relies on, it is characterised in that：View is cut into vertebra first and is cut to a shade Real space, this shade real space is then based on to create shadow map；It is concretely comprised the following steps：（1）, view cut vertebra cutThe non-shadow space that view is cut to vertebra part is deleted, so as to form a more close shadow object bounding box, by this Space after cutting is named as shade real space, and it is the minimum space for generating shadow map；（2）, shadow map generationFirstly generate light and cut vertebra, light section vertebra realizes the segmentation to shade visibility region, utilizes light to cut vertebra generation shade Figure.
- 2. the shadow map creation method that viewpoint as claimed in claim 1 relies on, it is characterised in that：The non-shadow space bag Include the space without projected objects on space and the horizontal plane under horizontal plane.
- 3. the shadow map creation method that viewpoint as claimed in claim 1 relies on, it is characterised in that：Using two-way partitioning algorithm View field is divided into multiple subregions, cuts vertebra according to these partition creating light successively.
- 4. the shadow map creation method that viewpoint as claimed in claim 1 relies on, it is characterised in that：Light cut vertebra viewport by Its respective partition determines, so the viewport of shadow map is also determined by its corresponding subregion.
- 5. the shadow map creation method that viewpoint as claimed in claim 1 relies on, it is characterised in that：Vertebra is cut for light, its Remote and hither plane should be by determining including all shadow objects inside its ken.
- 6. the shadow map creation method that viewpoint as claimed in claim 1 relies on, it is characterised in that：After light cuts vertebra determination, The object in light section vertebra, which is rendered, using shade camera obtains shadow map.
- 7. the shadow map creation method that viewpoint as claimed in claim 3 relies on, it is characterised in that：The two-way partitioning algorithm Including：It is longitudinally split：It is longitudinally split to view field's progress, farthest limit the wasting space of each subregion；Horizontal partition：For shadow map, its resolution ratioDifference should be limited to a certain limit, , wherein,WithThe respectively subregion of view fieldVertically and horizontally edge lengths；By being carried out again to vertical partition Laterally division, to meet this limitation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510255989.5A CN104952103B (en) | 2015-05-19 | 2015-05-19 | The shadow map creation method that a kind of viewpoint relies on |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510255989.5A CN104952103B (en) | 2015-05-19 | 2015-05-19 | The shadow map creation method that a kind of viewpoint relies on |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104952103A CN104952103A (en) | 2015-09-30 |
| CN104952103B true CN104952103B (en) | 2018-03-09 |
Family
ID=54166728
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510255989.5A Expired - Fee Related CN104952103B (en) | 2015-05-19 | 2015-05-19 | The shadow map creation method that a kind of viewpoint relies on |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104952103B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022002716A1 (en) * | 2020-06-30 | 2022-01-06 | Interdigital Ce Patent Holdings, Sas | Shadow-based estimation of 3d lighting parameters from reference object and reference virtual viewpoint |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106548506A (en) * | 2016-10-31 | 2017-03-29 | 中国能源建设集团江苏省电力设计院有限公司 | A kind of virtual scene Shading Rendering optimized algorithm based on layering VSM |
| CN111028357B (en) * | 2018-10-09 | 2020-11-17 | 北京嘀嘀无限科技发展有限公司 | Soft shadow processing method and device of augmented reality equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101393651A (en) * | 2008-11-07 | 2009-03-25 | 北京航空航天大学 | View field driving real-time shadow method |
| CN102436673A (en) * | 2011-10-24 | 2012-05-02 | 克拉玛依红有软件有限责任公司 | Shadow drafting method of large-scale outdoor scene |
| CN102903146A (en) * | 2012-09-13 | 2013-01-30 | 中国科学院自动化研究所 | Image processing method for scene drawing |
-
2015
- 2015-05-19 CN CN201510255989.5A patent/CN104952103B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101393651A (en) * | 2008-11-07 | 2009-03-25 | 北京航空航天大学 | View field driving real-time shadow method |
| CN102436673A (en) * | 2011-10-24 | 2012-05-02 | 克拉玛依红有软件有限责任公司 | Shadow drafting method of large-scale outdoor scene |
| CN102903146A (en) * | 2012-09-13 | 2013-01-30 | 中国科学院自动化研究所 | Image processing method for scene drawing |
Non-Patent Citations (3)
| Title |
|---|
| Parallel-Split Shadow Maps for Large-scale Virtual Environments;Zhang F 等;《Vrcia 2006 ACM International Coference on Virtual Reality Continuum and ITS Applications》;20061231;第311-318页 * |
| 航天器在轨飞行的实时可视化仿真;孙毅 等;《系统仿真学报》;20091031;第21卷;第124-127页 * |
| 视景仿真系统中实时阴影绘制技术的研究;张勇 等;《系统仿真学报》;20060831;第18卷;第180-182页 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022002716A1 (en) * | 2020-06-30 | 2022-01-06 | Interdigital Ce Patent Holdings, Sas | Shadow-based estimation of 3d lighting parameters from reference object and reference virtual viewpoint |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104952103A (en) | 2015-09-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chaurasia et al. | Depth synthesis and local warps for plausible image-based navigation | |
| CN105247575B (en) | System and method for being superimposed two dimensional map data on three-dimensional scenic | |
| CN102592275B (en) | Virtual viewpoint rendering method | |
| CN104331918A (en) | Occlusion culling and acceleration method for drawing outdoor ground surface in real time based on depth map | |
| CN105224288B (en) | Binocular three-dimensional method for rendering graph and related system | |
| US9639971B2 (en) | Image processing apparatus and method for processing transparency information of drawing commands | |
| CN104952103B (en) | The shadow map creation method that a kind of viewpoint relies on | |
| US20140375633A1 (en) | Texture Blending Between View-Dependent Texture and Base Texture in a Geographic Information System | |
| CN102243768A (en) | Method for drawing stereo picture of three-dimensional virtual scene | |
| CN104299257A (en) | Outdoor-sunlight-based method for realizing real-time dynamic shadow | |
| CN102306401B (en) | Left/right-eye three-dimensional picture drawing method for three-dimensional (3D) virtual scene containing fuzzy reflection effect | |
| CN109461197B (en) | Cloud real-time drawing optimization method based on spherical UV and re-projection | |
| DuVall et al. | Compositing light field video using multiplane images | |
| Gu et al. | Ue4-nerf: Neural radiance field for real-time rendering of large-scale scene | |
| Wißmann et al. | Accelerated stereo rendering with hybrid reprojection-based rasterization and adaptive ray-tracing | |
| CN108765582A (en) | A kind of panoramic pictures display methods and equipment | |
| CN107689076A (en) | A kind of efficient rendering intent during the cutting for system of virtual operation | |
| KR101574422B1 (en) | A method for rendering speed and editing efficiency improvement through single view video representation of multi-view video | |
| Kolivand et al. | Anti-aliasing in image based shadow generation techniques: a comprehensive survey | |
| KR20090040515A (en) | Apparatus and method for image space based shadowing with adaptive traversal scheme | |
| Trapp et al. | Occlusion management techniques for the visualization of transportation networks in virtual 3D city models | |
| Eggert et al. | Multi-layer visualization of mobile mapping data | |
| JP2005173866A (en) | Three-dimensional image display device | |
| Kolivand et al. | Shadow mapping or shadow volume? | |
| Luo et al. | Spherical projective displacement mesh |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180309 Termination date: 20200519 |