CN107527322A - Rendering intent, device, engine and storage medium - Google Patents
Rendering intent, device, engine and storage medium Download PDFInfo
- Publication number
- CN107527322A CN107527322A CN201710890960.3A CN201710890960A CN107527322A CN 107527322 A CN107527322 A CN 107527322A CN 201710890960 A CN201710890960 A CN 201710890960A CN 107527322 A CN107527322 A CN 107527322A
- Authority
- CN
- China
- Prior art keywords
- resolution
- texture maps
- target
- distribution
- light sensation
- 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
- 238000009877 rendering Methods 0.000 title claims abstract description 64
- 238000003860 storage Methods 0.000 title claims abstract description 13
- 230000035807 sensation Effects 0.000 claims abstract description 120
- 238000009826 distribution Methods 0.000 claims abstract description 115
- 238000000034 method Methods 0.000 claims abstract description 24
- 239000000284 extract Substances 0.000 claims abstract description 6
- 230000015654 memory Effects 0.000 claims description 19
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 238000000605 extraction Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000013527 convolutional neural network Methods 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 10
- 230000004927 fusion Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 235000013399 edible fruits Nutrition 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000005286 illumination Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 239000004744 fabric Substances 0.000 description 3
- 230000001186 cumulative effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 206010021703 Indifference Diseases 0.000 description 1
- 238000000342 Monte Carlo simulation Methods 0.000 description 1
- 238000005267 amalgamation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000004043 dyeing Methods 0.000 description 1
- 238000012966 insertion method Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4053—Scaling of whole images or parts thereof, e.g. expanding or contracting based on super-resolution, i.e. the output image resolution being higher than the sensor resolution
- G06T3/4076—Scaling of whole images or parts thereof, e.g. expanding or contracting based on super-resolution, i.e. the output image resolution being higher than the sensor resolution using the original low-resolution images to iteratively correct the high-resolution images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T15/00—3D [Three Dimensional] image rendering
- G06T15/06—Ray-tracing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Graphics (AREA)
- Image Generation (AREA)
Abstract
The embodiment of the invention discloses a kind of rendering intent, device, engine and storage medium, the rendering intent includes：The low resolution for obtaining target scene renders figure, target resolution texture maps and low resolution texture maps, and the resolution ratio that the low resolution texture maps render figure with the low resolution is identical；Extract the low resolution and render light sensation changes in distribution information of the figure relative to the low resolution texture maps, generate light sensation changes in distribution figure；Superresolution restoration is carried out to the light sensation changes in distribution figure, generates target resolution light sensation changes in distribution figure, the target resolution light sensation changes in distribution figure is identical with the resolution ratio of the target resolution texture maps；The target resolution light sensation changes in distribution figure is merged with the target resolution texture maps, result is obtained and renders figure.Solve the problems, such as that the high-resolution result of generation of the prior art renders the time-consuming longer of figure, realize the technique effect for reducing render time while rendering effect is ensured.
Description
Technical field
The present embodiments relate to Rendering field, more particularly to a kind of rendering intent, device, engine and storage to be situated between
Matter.
Background technology
Figure can intuitively express various information, and capacity is big, is easy to people's acquisition.With computer soft and hardware
The development advanced by leaps and bounds, computer graphics have obtained rapid popularization in the application of industry-by-industry and goed deep into.
But generally require to be rendered with the figure of vivid effect, and prior art render it is time-consuming often with figure
As pixel is directly proportional.For example the monte carlo method based on ray tracing is the expansion to traditional biggest advantage of light track algorithm, by general
Rate statistics method approximation render equation, can support more Facing material effects and simulation global illumination effect, obtain compared with
Good rendering effect.Specifically, if the one or more positions addition light source of user in the scene, the color, bright of light source is set
After the attribute such as degree and lighting angle, then illumination render is carried out based on these settings.The mode of rendering is, according to light source attributes, to calculate
Brightness of each light source on each pixel influences, and so as to adjust the pixel value of the pixel, completes illumination render.
But the illumination render mode of Monte Carlo ray tracing for each pixel, it is necessary to calculate substantial amounts of sampled light
Line, render time is especially long, and the high-resolution that one resolution ratio of generation is 4000*3000, which renders figure, needs a few minutes, is approximately equal to
Resolution ratio is 4 times of render time of 2000*1500, therefore high-resolution renders the cost that renders of figure and can be in any more always, makes
User will be often accepted or rejected between quality in rendering speed and rendering.
The content of the invention
The embodiment of the present invention provides a kind of rendering intent, device, engine and storage medium, solves life of the prior art
Into high-resolution result render figure it is time-consuming longer the problem of.
In a first aspect, the embodiments of the invention provide a kind of rendering intent, including：
The low resolution for obtaining target scene renders figure, target resolution texture maps and low resolution texture maps, described low
The resolution ratio that resolution texture figure renders figure with the low resolution is identical, and the low resolution texture maps and the low resolution
Rate renders resolution ratio of the resolution ratio less than the target resolution texture maps of figure；
Extract the low resolution and render light sensation changes in distribution information of the figure relative to the low resolution texture maps, generate
Light sensation changes in distribution figure；
Superresolution restoration is carried out to the light sensation changes in distribution figure, generates target resolution light sensation changes in distribution figure, institute
It is identical with the resolution ratio of the target resolution texture maps to state target resolution light sensation changes in distribution figure；
The target resolution light sensation changes in distribution figure is merged with the target resolution texture maps, result is obtained and renders
Figure.
Further, the low resolution for obtaining target scene renders figure, including：
Based on default low resolution values, target scene is rendered based on biggest advantage of light track algorithm, generates low resolution wash with watercolours
Dye figure.
Further, obtaining low resolution texture maps includes：
Based on default low resolution values, low resolution texture maps are generated according to the target scene；Or
Based on goal-selling resolution value, target resolution texture maps are generated according to the target scene；
Rule is reduced based on default resolution ratio, reduces the resolution ratio of the target resolution texture maps, generates low resolution
Texture maps.
Further, before obtaining low resolution texture maps, in addition to：
Judge the goal-selling resolution value whether higher than default resolution threshold；
If so, then triggering generates low resolution texture maps；
If it is not, then target scene is rendered according to goal-selling resolution value, knot of the generation with target resolution
Fruit renders figure.
Further, the basis, which presets resolution ratio, reduces rule, reduces the resolution ratio of the target resolution texture maps, raw
Include into low resolution texture maps：
By nearest point sampling, the resolution ratio of the target resolution texture maps is reduced, generates low resolution texture maps.
Further, the extraction low resolution renders figure and is distributed relative to the light sensation of the low resolution texture maps and become
Change information, generate light sensation changes in distribution figure, including：
The pixel value of each pixel of figure and the low pixel value for differentiating texture maps respective pixel are rendered to the low resolution
Calculated according to setup algorithm rule, the light sensation distributed intelligence in figure is rendered using result of calculation as the low resolution, it is raw
Into light sensation changes in distribution figure.
Further, the setup algorithm rule is：
Ask for the low each pixel value for differentiating texture maps and setting offset and, the renewal low resolution texture maps
Each pixel value；
Ask for the low resolution and render each pixel value of figure and the low respective pixel for differentiating texture maps after renewal
The business of value, as result of calculation.
Further, it is described that superresolution restoration, generation target resolution light sensation point are carried out to the light sensation changes in distribution figure
Cloth variation diagram, including：
The light sensation changes in distribution figure is subjected to superresolution restoration based on depth convolutional neural networks, generation target is differentiated
Rate light sensation changes in distribution figure.
Further, the target resolution light sensation changes in distribution figure is merged with the target resolution texture maps, obtained
As a result figure is rendered, including：
Ask for the target differentiate each pixel values of texture maps and setting offset and, update the target resolution texture
Each pixel value of figure；The pixel value of each pixel of target resolution light sensation changes in distribution figure is cumulative or mutually take the institute after renewal
On the pixel value for stating target resolution texture maps corresponding pixel points, form result and render figure.
Further, methods described is performed by rendering engine.
Second aspect, the embodiment of the present invention additionally provide a kind of rendering device, including：
Image collection module, the low resolution for obtaining target scene render figure, target resolution texture maps and low point
Resolution texture maps, the resolution ratio that the low resolution texture maps render figure with the low resolution is identical, and the low resolution
Texture maps render resolution ratio of the resolution ratio less than the target resolution texture maps of figure with the low resolution；
Light sensation changes in distribution figure extraction module, figure is rendered relative to the low resolution line for extracting the low resolution
The light sensation changes in distribution information of figure is managed, generates light sensation changes in distribution figure；
Superresolution restoration module, for carrying out superresolution restoration, generation target point to the light sensation changes in distribution figure
Resolution light sensation changes in distribution figure, the target resolution light sensation changes in distribution figure and the resolution ratio of the target resolution texture maps
It is identical；
Image co-registration module, for by the target resolution light sensation changes in distribution figure and the target resolution texture maps
Fusion, obtains result and renders figure.
The third aspect, the embodiment of the present invention additionally provide a kind of rendering engine, including：
Display, memory, processor and storage on a memory and the computer program that can run on a processor, institute
Rendering intent as described in relation to the first aspect is realized when stating computing device described program, the display is used to show the result
Render figure.
Fourth aspect, the embodiment of the present invention additionally provide a kind of computer-readable recording medium, are stored thereon with computer
Program, rendering intent as described in relation to the first aspect is realized when the program is executed by processor.
The technical scheme of rendering intent provided in an embodiment of the present invention, by rendering figure to low resolution relative to low resolution
The light sensation changes in distribution figure of rate texture maps carries out superresolution restoration, makes the target resolution light sensation changes in distribution figure of generation have
With target resolution texture maps identical resolution ratio, then by target resolution light sensation changes in distribution figure and target resolution texture maps
Fusion generation result renders figure；Light sensation changes in distribution figure carries low resolution and renders light of the figure relative to low resolution texture maps
Feel distributed intelligence, after carrying out superresolution restoration to it, make the target resolution texture maps and target resolution light sensation point of generation
The fusion of cloth variation diagram can obtain result and render figure, result is rendered figure and carry high-resolution spatial cue, avoid straight
Connect to render target scene and render figure to generate target resolution, greatly reduce render time.
Brief description of the drawings
Technical scheme in order to illustrate the embodiments of the present invention more clearly, make required in being described below to embodiment
Accompanying drawing does one and simply introduced, it should be apparent that, drawings in the following description are some embodiments of the present invention, for this
For the those of ordinary skill of field, on the premise of not paying creative work, it can also be obtained according to these accompanying drawings other
Accompanying drawing.
Fig. 1 is the flow chart for the rendering intent that the embodiment of the present invention one provides；
Fig. 2 is the flow chart of the extracting method for the low resolution texture maps that the embodiment of the present invention two provides；
Fig. 3 is the flow chart for the rendering intent that the embodiment of the present invention three provides；
Fig. 4 A are that the low resolution field that the embodiment of the present invention three provides renders figure；
Fig. 4 B are the target resolution field texture maps that the embodiment of the present invention three provides；
Fig. 4 C are the low resolution texture maps that the embodiment of the present invention three provides；
Fig. 4 D are the light sensation changes in distribution figures that the embodiment of the present invention three provides；
Fig. 4 E are the target resolution light sensation changes in distribution figures that the embodiment of the present invention three provides；
Fig. 4 F are that the result that the embodiment of the present invention three provides renders figure；
Fig. 4 G are that the direct result that the embodiment of the present invention three provides renders figure；
Fig. 5 is the structured flowchart schematic diagram for the rendering device that the embodiment of the present invention four provides；
Fig. 6 is the structured flowchart schematic diagram for the rendering engine that the embodiment of the present invention five provides.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, hereinafter with reference to attached in the embodiment of the present invention
Figure, technical scheme is clearly and completely described by embodiment, it is clear that described embodiment is the present invention one
Section Example, rather than whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art are not doing
Go out under the premise of creative work the every other embodiment obtained, belong to the scope of protection of the invention.
Embodiment one
Fig. 1 is the flow chart for the rendering intent that the embodiment of the present invention one provides.The technical scheme of the present embodiment be applied to pair
Scene render the situation that generation renders image, and this method is performed by configuring software in smart machine or hardware, such as,
Performed by mobile phone, computer, PAD or other smart machines with graphics processing function.As shown in figure 1, this method is specific
Comprise the following steps：
S101. the low resolution for obtaining target scene renders figure, target resolution texture maps and low resolution texture maps, low
Resolution texture figure is identical with the resolution ratio that low resolution renders figure, and low resolution texture maps render point of figure with low resolution
Resolution is less than the resolution ratio of target resolution texture maps.
Target scene is the scene that user is desired with rendering, and it is the default low resolution values of basis that low resolution, which renders figure,
The low resolution for target scene render generation renders figure, and the present embodiment preferably passes through the wash with watercolours pixel-by-pixel such as biggest advantage of light track algorithm
Dyeing method, to target scene render generation low resolution and render figure；Target resolution texture maps are according to goal-selling point
Resolution value, target scene is generated as target resolution texture maps；Low resolution texture maps can according to default low resolution values,
Target scene is generated into low resolution texture maps；Or according to goal-selling resolution value, target scene is generated into target
Resolution texture figure, rule is then reduced according to default resolution ratio, reduces the resolution ratio of target resolution texture maps, generates low point
Resolution texture maps.
Target resolution value is higher than low resolution values in the present embodiment, and specific target resolution value is not limited
System, a relative value can be understood as, i.e., relative to the processing speed and the acceptable figure of user of image processing equipment
As a relative value of render time.Resolution ratio of the high-resolution texture maps than the low resolution texture maps after reducing resolution ratio
It is big.
The present embodiment preferably by nearest point sampling, reduces the resolution ratio of target resolution texture maps, generates low resolution
Texture maps.And the resolution ratio of the low resolution texture maps of generation is preferably the 1/16~3/4 of target resolution texture maps resolution ratio,
Specific resolution ratio reduces ratio and can selected according to the size of target resolution value, such as, target resolution value itself
Resolution ratio is lower, then resolution ratio is bigger.Certainly, it will be understood by those skilled in the art that other resolution ratio can also be used
Rule or algorithm are reduced to realize.
S102. extraction low resolution renders light sensation changes in distribution information of the figure relative to low resolution texture maps, generates light
Feel changes in distribution figure.
Low resolution renders in figure and mainly extracts low resolution containing the spatial cue after rendering, the operation and render figure phase
Distinguished for the pixel of low resolution texture maps, can be the difference pixel value for obtaining each pixel one by one, one can also be passed through
A little algorithms obtain the value differences of compartmentalization.
S103. superresolution restoration is carried out to light sensation changes in distribution figure, generates target resolution light sensation changes in distribution figure, mesh
It is identical with the resolution ratio of target resolution texture maps to mark resolution ratio light sensation changes in distribution figure.
In order that the resolution ratio of light sensation changes in distribution figure is consistent with the resolution ratio of target resolution texture maps, the present embodiment pair
Light sensation changes in distribution figure carries out superresolution restoration, makes the target resolution light sensation changes in distribution figure and target resolution line of generation
The resolution ratio of reason figure is identical, so that point of the resolution ratio of target resolution light sensation changes in distribution figure and target resolution texture maps
Resolution has one-to-one relation.
The present embodiment is preferably based on depth convolutional neural networks and light sensation changes in distribution figure is carried out into superresolution restoration, generation
Target resolution light sensation changes in distribution figure.Using what is learnt in advance, the light sensation distribution of low resolution becomes depth convolutional neural networks
Change figure to the transformational relation of target resolution light sensation changes in distribution figure, superresolution restoration is carried out to light sensation changes in distribution figure.Phase
Than traditional interpolation algorithm (for example, calculating the average pixel value of neighbor pixel, the new pixel inserted between neighbor pixel
Point has average pixel value), the superresolution restoration method based on depth convolutional neural networks is relatively more complicated pixel
Point insertion method, can obtain more natural target resolution light sensation changes in distribution figure.
S104. target resolution light sensation changes in distribution figure is merged with target resolution texture maps, obtains result and render figure.
Aforesaid operations enter the target resolution light sensation changes in distribution figure with spatial cue with target resolution texture maps
Row fusion, so as to which spatial cue is added in target resolution texture maps.Specific amalgamation mode is optional：Ask for target resolution
Each pixel values of texture maps and setting offset and, each pixel value of more fresh target resolution texture maps；By target resolution light
The pixel value of sense each pixel of changes in distribution figure is cumulative or mutually takes the target resolution texture maps corresponding pixel points after renewal
On pixel value, form result and render figure.
Because the resolution ratio of target resolution light sensation changes in distribution figure and the resolution ratio of target resolution texture maps have one
One-to-one correspondence, therefore result renders light sensation distributed intelligence in figure and a pair of resolution ratio 1 in target resolution texture maps
Should, and then make it have, the result with directly carrying out rendering generation target resolution to target scene render figure as effect,
And the present embodiment renders generation low resolution to target scene and renders figure, then extracts the low resolution and renders figure and low resolution
The light sensation changes in distribution figure of rate texture maps, superresolution restoration generation target resolution is then carried out to the light sensation changes in distribution figure
Light sensation changes in distribution figure, and the time that target resolution light sensation changes in distribution figure is merged with target resolution texture maps, far
Much smaller than the time that the result for directly to target scene render generation target resolution renders figure, therefore, the application is protecting
While card result renders the effect of figure, render time is also greatly reduced.
The technical scheme of rendering intent provided in an embodiment of the present invention, by rendering figure to low resolution relative to low resolution
The light sensation changes in distribution figure of rate texture maps carries out superresolution restoration, makes the target resolution light sensation changes in distribution figure of generation have
With target resolution texture maps identical resolution ratio, then by target resolution light sensation changes in distribution figure and target resolution texture maps
Fusion generation result renders figure, and light sensation changes in distribution figure carries low resolution and renders light of the figure relative to low resolution texture maps
Feel distributed intelligence, and then make target resolution texture maps merge with target resolution light sensation changes in distribution figure to obtain result wash with watercolours
Dye figure, makes result render figure and carries high-resolution spatial cue, avoids and directly target scene is rendered to generate
Target resolution renders figure, greatly reduces render time.
Embodiment two
Fig. 2 is the flow of the extracting method of the low resolution texture maps in the rendering intent that the embodiment of the present invention two provides
Figure.The present embodiment further before low resolution texture maps are obtained, is judged the goal-selling resolution value.Such as figure
Shown in 2, the low resolution texture picture capturing method includes：
S1001. whether goal-selling resolution value is judged higher than default resolution threshold, if then triggering performs S1002,
If otherwise perform S1003.
Default resolution threshold in the present embodiment can be set according to actual conditions, for example result renders the effect of figure
Fruit requires, performs the performance of rendering apparatus and the requirement of rendering speed etc..
S1002. triggering generation low resolution texture maps.
The low generating mode for differentiating texture maps in the present embodiment, including：According to default low resolution values, according to target scene
Generate low resolution texture maps；Or according to goal-selling resolution value, target resolution texture maps are generated according to target scene；
Rule is reduced according to default resolution ratio, reduces the resolution ratio of target resolution texture maps, generates low resolution texture maps.
S1003. according to goal-selling resolution value, target scene is rendered, knot of the generation with target resolution
Fruit renders figure.
When goal-selling resolution value is less than default resolution threshold, then explanation is directly rendered to target scene
Time in tolerance interval, now directly renders to target scene.
In the embodiment of the present invention, judge whether to need by the relation of goal-selling resolution value and default resolution threshold
Generate low resolution texture maps, if desired, then generate low resolution texture maps, and pass through the rendering intent described in above-described embodiment
Generation renders figure with the result of target resolution, if need not if directly target scene is rendered, generation has target
The result of resolution ratio renders figure, facilitates user to set default resolution threshold according to the actual requirements, and practicality is higher.
Embodiment three
Fig. 3 is the flow chart for the rendering intent that the embodiment of the present invention three provides.The embodiment of the present invention is in above-described embodiment
On the basis of, light sensation changes in distribution information of the figure relative to low resolution texture maps is rendered to extracting the low resolution, generates light
Feel the optimization of changes in distribution figure.As shown in figure 3, the rendering intent includes：
S101. the low resolution for obtaining target scene renders figure, target resolution texture maps and low resolution texture maps, low
Resolution texture figure is identical with the resolution ratio that low resolution renders figure, and low resolution texture maps render point of figure with low resolution
Resolution is less than the resolution ratio of target resolution texture maps.
S102. low resolution is rendered each pixel of figure pixel value and it is low differentiate texture maps respective pixel pixel value according to
Setup algorithm rule is calculated, and the light sensation distributed intelligence in figure, generation light sensation point are rendered using result of calculation as low resolution
Cloth variation diagram.
The setup algorithm rule is：Ask for it is described it is low differentiate texture maps each pixel value with setting offset and, more
The new low each pixel value for differentiating texture maps；Asking for the low resolution, to render each pixel value of figure described low with after renewal
The business of the respective pixel value of texture maps is differentiated, as result of calculation..Or can also be the difference for seeking pixel value, or pixel value
Between other computation rules, as long as pixel value variation can be embodied.
The present embodiment is preferably but not limited to directly render figure divided by low resolution texture maps by low resolution to extract low point
Resolution renders the light sensation distributed intelligence in figure, so as to generate light sensation changes in distribution figure.
S103. superresolution restoration is carried out to light sensation changes in distribution figure, generates target resolution light sensation changes in distribution figure, mesh
It is identical with the resolution ratio of target resolution texture maps to mark resolution ratio light sensation changes in distribution figure.
S104. target resolution light sensation changes in distribution figure is merged with target resolution texture maps, obtains result and render figure.
Further, target scene is rendered in the present embodiment and renders figure to generate low resolution, to target field
Scape can carry out a small amount of sampling when being rendered to each pixel, can also largely be sampled.Rendering result figure
The quality of effect is substantially directly proportional to sampling number.
Exemplary, when discovery result, which renders figure, does not meet expected require, pixel can be increased to target scene and adopted
Sample number, re-starts to render and renders figure to generate low resolution；Need to add new spatial cue when discovery result renders figure
When, it can dose to re-start after new light information in target scene and render, generation low resolution renders figure, with more new low point
Resolution ratio renders figure, is then based on that new low resolution renders figure and the rendering intent generation target described in previous embodiment is differentiated
The result of rate renders figure.
Exemplary, target scene is rendered according to default low resolution values, generation low resolution renders figure, takes
30s, resolution ratio 800*600, as shown in Figure 4 A；According to goal-selling resolution value, target is generated according to target scene and differentiated
Rate texture maps, 1.5s, target resolution 1600*1200 are taken, as shown in Figure 4 B；Reduce the resolution of target resolution texture maps
Rate generates low resolution texture maps, takes 0.5s, as shown in Figure 4 C to 800*600；Extraction low resolution renders figure relative to low
The light sensation changes in distribution information of resolution texture figure, light sensation changes in distribution figure is generated, take 0.1s, resolution ratio 800*600, such as
Shown in Fig. 4 D；Superresolution restoration is carried out to light sensation changes in distribution figure, generates target resolution light sensation changes in distribution figure, is taken
3s, resolution ratio 1600*1200, as shown in Figure 4 E, by target resolution light sensation changes in distribution figure and target resolution texture maps
Fusion, obtains result and renders figure, takes 0.1s, resolution ratio 1600*1200, and as illustrated in figure 4f, it is 35.2s to take summation.
According to goal-selling resolution value, directly target scene is rendered, result of the generation with target resolution
Figure is rendered, i.e. resolution ratio is that 1600*1200 result renders figure, takes 150s, as shown in Figure 4 G.
Comparison diagram 4F and Fig. 4 G, the two visually and indifference, but based on rendering intent of the present invention when
Between summation be 35.2s, far smaller than directly target scene is rendered in the prior art, generation with target resolution knot
Fruit renders the time of figure.
The present embodiment renders light sensation changes in distribution figure in figure by extracting low resolution, realizes that low resolution is rendered in figure
Light sensation distributed intelligence and the other information such as texture separation, so as to which light sensation changes in distribution figure and target resolution texture maps be melted
Close and realize the corresponding pixel-by-pixel of spatial cue and target resolution texture maps, result is rendered figure has, and based on prior art
While the result of generation renders figure the same visual effect, render time is reduced.
The rendering intent that various embodiments of the present invention are provided, is preferably performed by rendering engine.The interface of rendering engine obtains
Target scene to be rendered is taken, after rendering, then figure is rendered to result and shown.
Example IV
Fig. 5 is the structured flowchart schematic diagram for the rendering device that the embodiment of the present invention four provides.The rendering device is used to perform
The rendering intent that above-mentioned any embodiment is provided, and be configurable in smart machine.As shown in figure 5, the rendering device includes：
Image collection module 11, the low resolution for obtaining target scene render figure, target resolution texture maps and low
Resolution texture figure, the resolution ratio that the low resolution texture maps render figure with the low resolution is identical, and the low resolution
Rate texture maps render resolution ratio of the resolution ratio less than the target resolution texture maps of figure with the low resolution；
Light sensation changes in distribution figure extraction module 12, figure is rendered relative to the low resolution for extracting the low resolution
The light sensation changes in distribution information of texture maps, generate light sensation changes in distribution figure；
Superresolution restoration module 13, for carrying out superresolution restoration to the light sensation changes in distribution figure, generate target
Resolution ratio light sensation changes in distribution figure, the target resolution light sensation changes in distribution figure and the resolution of the target resolution texture maps
Rate is identical；
Image co-registration module 14, for by the target resolution light sensation changes in distribution figure and the target resolution texture
Figure fusion, obtains result and renders figure.
The technical scheme of rendering device provided in an embodiment of the present invention, by rendering figure to low resolution relative to low resolution
The light sensation changes in distribution figure of rate texture maps carries out superresolution restoration, makes the target resolution light sensation changes in distribution figure of generation have
With target resolution texture maps identical resolution ratio, then by target resolution light sensation changes in distribution figure and target resolution texture maps
Fusion generation result renders figure, and light sensation changes in distribution figure carries low resolution and renders light of the figure relative to low resolution texture maps
Feel distributed intelligence, and then make target resolution texture maps merge with target resolution light sensation changes in distribution figure to obtain result wash with watercolours
Dye figure, makes result render figure and carries high-resolution spatial cue, avoids and directly target scene is rendered to generate
Target resolution renders figure, greatly reduces render time.
The rendering device that the embodiment of the present invention is provided can perform the rendering intent that any embodiment of the present invention is provided, tool
For the corresponding functional module of execution method and beneficial effect.
Embodiment five
Fig. 6 is a kind of structural representation for rendering engine that the embodiment of the present invention five provides, as shown in fig. 6, the equipment bag
Include display 300, processor 301, memory 302；The quantity of processor 301 can be one or more in engine, in Fig. 6 with
Exemplified by one processor 301；Display, processor 301, memory 302 in equipment can be connected by bus or other modes
Connect, in Fig. 6 exemplified by being connected by bus.
Memory 302 is used as a kind of computer-readable recording medium, and journey is can perform available for storage software program, computer
Sequence and module, as corresponding to the rendering intent in the embodiment of the present invention programmed instruction/module (for example, image collection module 11,
Light sensation changes in distribution figure extraction module 12, superresolution restoration module 13 and image co-registration module 14).Processor 301 passes through
Software program, instruction and the module of operation storage in the memory 302, so as to perform the various function application of equipment and number
According to processing, that is, realize above-mentioned rendering intent.
Memory 302 can mainly include storing program area and storage data field, wherein, storing program area can store operation system
Application program needed for system, at least one function；Storage data field can store uses created data etc. according to terminal.This
Outside, memory 302 can include high-speed random access memory, can also include nonvolatile memory, for example, at least one
Disk memory, flush memory device or other non-volatile solid state memory parts.In some instances, memory 302 can enter one
Step includes that relative to the remotely located memory of processor 301, these remote memories network connection to equipment can be passed through.On
The example for stating network includes but is not limited to internet, intranet, LAN, mobile radio communication and combinations thereof.
Display 300 may include the display devices such as display screen, for example, the display screen of user terminal.
Embodiment six
The embodiment of the present invention six also provides a kind of storage medium for including computer executable instructions, and the computer can be held
Row instruction is used to perform a kind of rendering intent when being performed by computer processor, and this method includes：
The low resolution for obtaining target scene renders figure, target resolution texture maps and low resolution texture maps, described low
The resolution ratio that resolution texture figure renders figure with the low resolution is identical, and the low resolution texture maps and the low resolution
Rate renders resolution ratio of the resolution ratio less than the target resolution texture maps of figure；
Extract the low resolution and render light sensation changes in distribution information of the figure relative to the low resolution texture maps, generate
Light sensation changes in distribution figure；
Superresolution restoration is carried out to the light sensation changes in distribution figure, generates target resolution light sensation changes in distribution figure, institute
It is identical with the resolution ratio of the target resolution texture maps to state target resolution light sensation changes in distribution figure；
The target resolution light sensation changes in distribution figure is merged with the target resolution texture maps, result is obtained and renders
Figure.
Certainly, a kind of storage medium for including computer executable instructions that the embodiment of the present invention is provided, its computer
The method operation that executable instruction is not limited to the described above, can also carry out the rendering intent that any embodiment of the present invention is provided
In associative operation.
By the description above with respect to embodiment, it is apparent to those skilled in the art that, the present invention
It can be realized by software and required common hardware, naturally it is also possible to realized by hardware, but the former is more in many cases
Good embodiment.Based on such understanding, what technical scheme substantially contributed to prior art in other words
Part can be embodied in the form of software product, and the computer software product can be stored in computer-readable recording medium
In, floppy disk, read-only storage (Read-Only Memory, abbreviation ROM), random access memory (Random such as computer
Access Memory, abbreviation RAM), flash memory (FLASH), hard disk or CD etc., including some instructions are causing a calculating
Machine equipment (can be personal computer, server, or network equipment etc.) performs rendering described in each embodiment of the present invention
Method.
It is worth noting that, in the embodiment of above-mentioned rendering device, included unit and module are simply according to work(
Energy logic is divided, but is not limited to above-mentioned division, as long as corresponding function can be realized；In addition, each work(
The specific name of energy unit is also only to facilitate mutually distinguish, the protection domain being not intended to limit the invention.
Pay attention to, above are only presently preferred embodiments of the present invention and institute's application technology principle.It will be appreciated by those skilled in the art that
The invention is not restricted to specific embodiment described here, can carry out for a person skilled in the art various obvious changes,
Readjust and substitute without departing from protection scope of the present invention.Therefore, although being carried out by above example to the present invention
It is described in further detail, but the present invention is not limited only to above example, without departing from the inventive concept, also
Other more equivalent embodiments can be included, and the scope of the present invention is determined by scope of the appended claims.
Claims (13)
- A kind of 1. rendering intent, it is characterised in that including：The low resolution for obtaining target scene renders figure, target resolution texture maps and low resolution texture maps, the low resolution The resolution ratio that rate texture maps render figure with the low resolution is identical, and the low resolution texture maps and the low resolution wash with watercolours The resolution ratio for contaminating figure is less than the resolution ratio of the target resolution texture maps；Extract the low resolution and render light sensation changes in distribution information of the figure relative to the low resolution texture maps, generate light sensation Changes in distribution figure；Superresolution restoration is carried out to the light sensation changes in distribution figure, generates target resolution light sensation changes in distribution figure, the mesh It is identical with the resolution ratio of the target resolution texture maps to mark resolution ratio light sensation changes in distribution figure；The target resolution light sensation changes in distribution figure is merged with the target resolution texture maps, result is obtained and renders figure.
- 2. according to the method for claim 1, it is characterised in that the low resolution for obtaining target scene renders figure, wraps Include：Based on default low resolution values, target scene is rendered according to biggest advantage of light track algorithm, generation low resolution renders figure.
- 3. according to the method for claim 1, it is characterised in that obtaining low resolution texture maps includes：Based on default low resolution values, low resolution texture maps are generated according to the target scene；OrBased on goal-selling resolution value, target resolution texture maps are generated according to the target scene；Rule is reduced based on default resolution ratio, reduces the resolution ratio of the target resolution texture maps, generates low resolution texture Figure.
- 4. according to the method for claim 3, it is characterised in that before obtaining low resolution texture maps, in addition to：Judge the goal-selling resolution value whether higher than default resolution threshold；If so, then triggering generates low resolution texture maps；If it is not, then target scene is rendered according to goal-selling resolution value, result wash with watercolours of the generation with target resolution Dye figure.
- 5. according to the method for claim 3, it is characterised in that it is described regular based on default resolution ratio reduction, described in reduction The resolution ratio of target resolution texture maps, generation low resolution texture maps include：By nearest point sampling, the resolution ratio of the target resolution texture maps is reduced, generates low resolution texture maps.
- 6. according to the method for claim 1, it is characterised in that the extraction low resolution renders figure relative to described The light sensation changes in distribution information of low resolution texture maps, light sensation changes in distribution figure is generated, including：The low resolution is rendered each pixel of figure pixel value and it is described it is low differentiate texture maps respective pixel pixel value according to Setup algorithm rule is calculated, and the light sensation distributed intelligence in figure is rendered using result of calculation as the low resolution, generates light Feel changes in distribution figure.
- 7. according to the method for claim 6, it is characterised in that the setup algorithm rule is：Ask for it is described it is low differentiate texture maps each pixel value with setting offset and, renewal it is described it is low resolution texture maps each picture Element value；Ask for the low resolution and render each pixel value of figure and the low respective pixel value for differentiating texture maps after renewal Business, as result of calculation.
- 8. according to the method for claim 1, it is characterised in that described that super-resolution is carried out to the light sensation changes in distribution figure Recover, generate target resolution light sensation changes in distribution figure, including：The light sensation changes in distribution figure is subjected to superresolution restoration based on depth convolutional neural networks, generates target resolution light Feel changes in distribution figure.
- 9. according to the method for claim 1, it is characterised in that by the target resolution light sensation changes in distribution figure with it is described Target resolution texture maps merge, and obtain result and render figure, including：Ask for the target differentiate each pixel values of texture maps and setting offset and, to update the target resolution texture maps Each pixel value；The pixel value of each pixel of target resolution light sensation changes in distribution figure is added up or mutually takes the target after renewal and is divided On the pixel value of resolution texture maps corresponding pixel points, form result and render figure.
- 10. according to any described methods of claim 1-9, it is characterised in that methods described is performed by rendering engine.
- A kind of 11. rendering device, it is characterised in that including：Image collection module, the low resolution for obtaining target scene render figure, target resolution texture maps and low resolution Texture maps, the resolution ratio that the low resolution texture maps render figure with the low resolution is identical, and the low resolution texture Figure renders resolution ratio of the resolution ratio less than the target resolution texture maps of figure with the low resolution；Light sensation changes in distribution figure extraction module, figure is rendered relative to the low resolution texture maps for extracting the low resolution Light sensation changes in distribution information, generate light sensation changes in distribution figure；Superresolution restoration module, for carrying out superresolution restoration to the light sensation changes in distribution figure, generate target resolution Light sensation changes in distribution figure, the target resolution light sensation changes in distribution figure and the resolution ratio phase of the target resolution texture maps Together；Image co-registration module, for the target resolution light sensation changes in distribution figure and the target resolution texture maps to be melted Close, obtain result and render figure.
- 12. a kind of rendering engine, including display, memory, processor and storage can transport on a memory and on a processor Capable computer program, it is characterised in that realized during the computing device described program as described in claim 1~10 is any Rendering intent, the display be used for show that the result renders figure.
- 13. a kind of computer-readable recording medium, is stored thereon with computer program, it is characterised in that the program is by processor The rendering intent as described in claim 1~10 is any is realized during execution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710890960.3A CN107527322B (en) | 2017-09-27 | 2017-09-27 | Rendering method, device, engine and storage medium combined with convolutional neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710890960.3A CN107527322B (en) | 2017-09-27 | 2017-09-27 | Rendering method, device, engine and storage medium combined with convolutional neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107527322A true CN107527322A (en) | 2017-12-29 |
| CN107527322B CN107527322B (en) | 2020-08-04 |
Family
ID=60737624
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710890960.3A Active CN107527322B (en) | 2017-09-27 | 2017-09-27 | Rendering method, device, engine and storage medium combined with convolutional neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107527322B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110738626A (en) * | 2019-10-24 | 2020-01-31 | 广东三维家信息科技有限公司 | Rendering graph optimization method and device and electronic equipment |
| CN111696188A (en) * | 2020-04-26 | 2020-09-22 | 杭州群核信息技术有限公司 | Rendering graph rapid illumination editing method and device and rendering method |
| CN112419467A (en) * | 2020-11-05 | 2021-02-26 | 杭州群核信息技术有限公司 | Rendering efficiency improving method, device and system based on deep learning |
| CN112801878A (en) * | 2021-02-08 | 2021-05-14 | 广东三维家信息科技有限公司 | Rendering image super-resolution texture enhancement method, device, equipment and storage medium |
| CN112905293A (en) * | 2021-03-26 | 2021-06-04 | 北京房江湖科技有限公司 | Graph loading method and system and graph rendering method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1584667A (en) * | 2004-06-15 | 2005-02-23 | 南京大学 | Self adaptable photon density controlling method driven by importance grade |
| CN105453139A (en) * | 2013-07-17 | 2016-03-30 | 微软技术许可有限责任公司 | Sparse GPU voxelization for 3D surface reconstruction |
| CN105676470A (en) * | 2016-03-24 | 2016-06-15 | 清华大学 | 3D scene vision spatial resolution enhancing method and system |
| CN106127684A (en) * | 2016-06-22 | 2016-11-16 | 中国科学院自动化研究所 | Image super-resolution Enhancement Method based on forward-backward recutrnce convolutional neural networks |
-
2017
- 2017-09-27 CN CN201710890960.3A patent/CN107527322B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1584667A (en) * | 2004-06-15 | 2005-02-23 | 南京大学 | Self adaptable photon density controlling method driven by importance grade |
| CN105453139A (en) * | 2013-07-17 | 2016-03-30 | 微软技术许可有限责任公司 | Sparse GPU voxelization for 3D surface reconstruction |
| CN105676470A (en) * | 2016-03-24 | 2016-06-15 | 清华大学 | 3D scene vision spatial resolution enhancing method and system |
| CN106127684A (en) * | 2016-06-22 | 2016-11-16 | 中国科学院自动化研究所 | Image super-resolution Enhancement Method based on forward-backward recutrnce convolutional neural networks |
Non-Patent Citations (2)
| Title |
|---|
| 李佳宁 等: "基于体分割重建的增强现实交互系统", 《光电子技术》 * |
| 肖春霞: "基于联合双边滤波的纹理合成上采样算法", 《计算机学报》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110738626A (en) * | 2019-10-24 | 2020-01-31 | 广东三维家信息科技有限公司 | Rendering graph optimization method and device and electronic equipment |
| CN110738626B (en) * | 2019-10-24 | 2022-06-28 | 广东三维家信息科技有限公司 | Rendering graph optimization method and device and electronic equipment |
| CN111696188A (en) * | 2020-04-26 | 2020-09-22 | 杭州群核信息技术有限公司 | Rendering graph rapid illumination editing method and device and rendering method |
| CN111696188B (en) * | 2020-04-26 | 2023-10-03 | 杭州群核信息技术有限公司 | Rendering graph rapid illumination editing method and device and rendering method |
| CN112419467A (en) * | 2020-11-05 | 2021-02-26 | 杭州群核信息技术有限公司 | Rendering efficiency improving method, device and system based on deep learning |
| CN112419467B (en) * | 2020-11-05 | 2023-10-03 | 杭州群核信息技术有限公司 | Method, device and system for improving rendering efficiency based on deep learning |
| CN112801878A (en) * | 2021-02-08 | 2021-05-14 | 广东三维家信息科技有限公司 | Rendering image super-resolution texture enhancement method, device, equipment and storage medium |
| CN112905293A (en) * | 2021-03-26 | 2021-06-04 | 北京房江湖科技有限公司 | Graph loading method and system and graph rendering method and system |
| CN112905293B (en) * | 2021-03-26 | 2023-07-07 | 贝壳找房(北京)科技有限公司 | Graphics loading method and system and graphics rendering method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107527322B (en) | 2020-08-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107527322A (en) | Rendering intent, device, engine and storage medium | |
| CN107680042A (en) | Rendering intent, device, engine and storage medium | |
| CN107742317A (en) | A kind of acquisition methods for rendering image, device, rendering system and storage medium | |
| CN107025457A (en) | A kind of image processing method and device | |
| JP6970283B2 (en) | Image stitching method and device, storage medium | |
| US8687015B2 (en) | Brushing tools for digital image adjustments | |
| CN105138317A (en) | Window display processing method and device applied to terminal equipment | |
| CN101606179B (en) | Universal front end for masks, selections and paths | |
| JP6378323B2 (en) | Image editing transmission to subordinate video sequences via dense motion fields | |
| CN106997608A (en) | A kind of method and device for generating halation result figure | |
| US7778486B2 (en) | Online image processing systems and methods | |
| DE102020001165A1 (en) | Automatic synthesis of a content-sensitive removal area for a content-sensitive filling | |
| US10628984B2 (en) | Facial model editing method and apparatus | |
| CN113763526B (en) | Hair highlight rendering method, device, equipment and storage medium | |
| CN110874816B (en) | Image processing method, device, mobile terminal and storage medium | |
| CN113256484A (en) | Method and device for stylizing image | |
| KR101189687B1 (en) | Method for creating 3d character | |
| Liu et al. | Video cloning for paintings via artistic style transfer | |
| CN110620961A (en) | Video watermark adding method and device and video editing terminal | |
| CN110457623A (en) | Acquisition methods, server and the storage medium of webpage frame | |
| CN114638951B (en) | House model display method and device, electronic equipment and readable storage medium | |
| CN107509101A (en) | Image processing method | |
| CN109003225A (en) | A kind of more palace trrellis diagram piece treating method and apparatus and a kind of electronic equipment | |
| CN105868674A (en) | Two-dimensional code scanning processing method and apparatus | |
| US20120120197A1 (en) | Apparatus and method for sharing hardware between graphics and lens distortion operation to generate pseudo 3d display |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Ye Qing Inventor after: Huang Xiaohuang Inventor after: Tang Rui Inventor after: Zhang Junfei Inventor after: Huang Yuzhong Inventor before: Ye Qing Inventor before: Tang Rui Inventor before: Zhang Junfei Inventor before: Huang Yuzhong |