CN104715495B - Three-dimensional model transmitting method based on user's specification error precision - Google Patents
Three-dimensional model transmitting method based on user's specification error precision Download PDFInfo
- Publication number
- CN104715495B CN104715495B CN201510057252.2A CN201510057252A CN104715495B CN 104715495 B CN104715495 B CN 104715495B CN 201510057252 A CN201510057252 A CN 201510057252A CN 104715495 B CN104715495 B CN 104715495B
- Authority
- CN
- China
- Prior art keywords
- coordinates
- model
- low frequency
- user
- frequency aberration
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
The invention discloses a kind of three-dimensional model transmitting method based on user's specification error precision.The present invention allows user in the collimation error and low frequency aberration of the specified required mode of client first.Secondly the geometric data of the model is done Laplacian conversions by service end, and the coordinate after conversion is referred to as δ coordinates coordinates.Then, the collimation error selected according to user, corresponding quantization is done by δ coordinates coordinates.Pass through Selecting All Parameters againThe anchor point of two different values, come tectonic model low frequency aberration and parameterBetween functional relation, and the low frequency aberration specified according to this function and user chooses the anchor point quantity added needed for the model.Finally the low frequency aberration of the model is finely adjusted, it is met the requirement of user, and the model is sent into client.According to the inventive method, user can be selected according to the demand of current network quality and reality needed for transmission threedimensional model quality.
Description
Technical field
The invention belongs to the multimedia technology field related to network transmission, more particularly to a kind of user's specification error precision
Threedimensional model compression transmitting method.
Background technology
With the development and the industrially extensive use to threedimensional model of computer graphics, threedimensional model is just towards multiple
Hydridization, the direction to become more meticulous are developed.Pass through the dimensional Modeling Technology of maturation, some large-scale virtual scenes such as 3D museums, three
Dimension virtual city, landform lifelike can be presented in us at the moment.These scenes are generally in large scale, dough sheet quantity
Numerous, high resolution, visual effect is splendid.Therefore whole model is extremely complex and model data amount is big.In addition, in modernization
In industrial design, the structure of industrial design product becomes increasingly complex, and the parts of composition are more and more, and each size parts
It is required for being expressed by threedimensional model.Therefore, industrial product structure is more complicated, and the structure of threedimensional model is also more complicated.
Another approach for obtaining threedimensional model be by 3-D scanning technology, with the continuous improvement of scanning technique,
It is more and more lower that the cost of threedimensional model is obtained by scanning, and the precision more and more higher of model.For some true, complexity
Its local detail feature rich in kind, the curvature of surface mesh is big, such as the face of human body, the art work etc., if being built by three-dimensional
Mould approach obtains its threedimensional model, and the manufacture craft of art designing will be extremely huge, often seem unable to do what one wishes, and be swept by three-dimensional
Technology is retouched, easily the space coordinates of object, color-values can be swept into computer rapidly, is obtained and natural scale, shape, face
The same threedimensional model of color, operating efficiency are improved.In ideo display stunt production field, 3-D scanning technology extensively should
For the entertainment medium industry such as film and tv industry, advertising, special effects with having brought unprecedented visual enjoyment,
The criterion of high quality production of film and TV is turned into.Show each details of material object, 3-D scanning skill in order to ultra high-definition
Art just develops towards the direction to become more meticulous.
Because the scale of threedimensional model is more and more huger, mould shapes become increasingly complex, and the precision more and more higher of model, lead
The data volume of processing and transmission needed for threedimensional model is caused to sharply increase.At the same time, it is general with PC and intelligent movable equipment
And increasing domestic consumer to the browsing of threedimensional model, the demand such as share and also sharply increase.Especially led in ecommerce
Domain, in order to attract customer, businessman needs to show the internal information of commodity in all directions, such as structure, function information, allows shopper's energy
It is enough to be roamed in dummy market, threedimensional model is operated with interacting formula.
In three-dimensional model transmitting, by being compressed processing to the data volume of threedimensional model, can significantly it reduce
The data volume of required transmission, so as to save limited network bandwidth and reduce the data processing amount of client.According to network matter
To the difference of threedimensional model demand, the multiresolution progressive compression transmission method of threedimensional model is wherein by the difference of amount and user
A solution.But, it is that optimal middle drafting effect, under given bit rate, progressive pressure are realized in client
Contracting transmission method need to be weighed between the two in the compression ratio and drafting distortion rate of model, and this calculating need to expend largely
Calculation cost.In addition, in some Mobile solutions, as moving game and mobile virtual roam, the threedimensional model in complex scene
The compression method of single bit rate is generally all used, so that threedimensional model quickly can be transferred to client from service end, and
There is higher compression ratio using the compression method of single resolution ratio.
The content of the invention
The present invention is directed to deficiency of the prior art on single resolution model Compression Transmission Technology, it is proposed that one kind is based on use
The three-dimensional model transmitting method of family specification error precision.
The technical solution adopted for the present invention to solve the technical problems is as follows：
First, user is allowed to specify the collimation error and low frequency aberration of required mode in client.Secondly, service end will
The geometric data (cartesian coordinate) of the model does Laplacian conversions, and the coordinate after conversion is referred to as δ-coordinates seats
Mark.Then, the collimation error selected according to user, corresponding quantization is done by δ-coordinates coordinates.Pass through Selecting All Parameters again
The anchor point of υ (parameter υ represents anchor point dense degree) two different values, the function come between tectonic model low frequency aberration and parameter υ
Relation, and the low frequency aberration specified according to this function and user chooses the anchor point quantity added needed for the model.Finally, to this
The low frequency aberration of model is finely adjusted, and it is met the requirement of user, and the model is sent into client.Client receives number
According to rear, a demand solution least square solution, the geometric data with regard to the model can be recovered.
Beneficial effects of the present invention：The present invention is a kind of threedimensional model compression transmission side based on user's specification error precision
Method, user can be according to the demands of current network quality and reality come the threedimensional model quality of transmission needed for selecting.
Brief description of the drawings
Fig. 1 is anchor point distribution and path.
Embodiment
Below in conjunction with accompanying drawing, the invention will be further described：
The present invention allows user in the collimation error and low frequency aberration of the specified required mode of client first.
Collimation error review extraction is：
Wherein：
L in above-mentioned formulaijSummit i to j distance is represented, N (i) represents the set of the abutment points on i summits, Q (vi) it is reconstruction model
Corresponding vertex vi。
Low frequency aberration review extraction is：
Wherein：viFor original geometric coordinate, Q (vi) be reconstruction model geometric coordinate.
Service end upon receiving a request, will do following processing：
Step 1：The geometric data of threedimensional model is converted into relative coordinate form from cartesian coordinate form.Define V=
{v1,v2,,,vnBe grid M vertex set, wherein vi=(xi,yi,zi).δ-coordinates the coordinates for defining each summit areWherein diThe number (being the degree on summit) abutted directly against a little on summit is represented,For xi,yi,ziCorresponding δ-coordinates coordinates.It is L=D-A to define Laplacian matrixes, based on wherein D
Diagonal matrix, the element on leading diagonal are：Dii=di, A is grid M adjacency matrix.It has following relation：Lx=D δ(x),
Ly=D δ(y), Lz=D δ(z).The x of grid, y, z coordinate can be converted into by δ-coordinates coordinates by three above relational expression.
Step 2：δ-coordinates coordinates are quantified.By the cartesian coordinate of threedimensional model be converted into δ-
After coordinates relative coordinates, real-coded GA can not be directly with coding or entropy code based on dictionary, and must pass through
Quantify.The visual level of the required mode selected according to client, select corresponding bits to quantify in service end, quantify model
Enclose for 3-8bits.
Step 3：Increment anchor point based on BFS (breadth first traversal) is chosen.Addition anchor point can reduce the low of reconstruction model
Frequency error, BFS anchor point selection process are as follows：A summit is randomly selected first as anchor point, and root is then used as using this point
Node, breadth first traversal (BFS) is done to the topology of model, records the depth on each layer of summit, that is, apart from root node (anchor
Point) topology distance, select that most deep point of depth as the 2nd anchor point as.Again using the 2nd anchor point as root node, start to do extensively
First traversal is spent, if the depth recorded before current depth ratio is small, changes it.It is selected before when most deep depth υ=15
All anchor points be put into container (υ15).Continue to select anchor point, until most deep depth is υ=14, anchor point number during and υ=14 is υ
Anchor point when=15 additional anchor point Δ newly increased again14, the anchor point that this part newly increases is put into container (Δ14).Class according to this
Push away, until anchor point increment during υ=5 is put into container (Δ5).Fig. 1 show anchor point distribution during parameter υ=3.Black color dots are anchor
Point, white point are general point, dotted line be in figure non-anchor apart from the maximum path value of its nearest anchor point.
Step 4：Construct reconstruction model low frequency aberration MqFunctional relation M between parameter υq=f (v), its by δ-
The difference of coordinates coordinate quantified precisions can be divided into three kinds of situations, and the low frequency aberration precision selected according to user is obtained
Required anchor point quantity.
Wherein function Mq=f (v) three kinds of concrete conditions are as follows：
1) if δ-coordinates coordinates are quantified by 3bits, 4bits or 5bits, now low frequency aberration MqWith parameter υ it
Between function Mq=f (v) is linear：Mq=av+b.Selecting All Parameters υ 2 different values, by solving least square method weight
Structure model, respectively obtain reconstruction model error amount Mq1,Mq2.By these two pair value ((υ1,Mq1), (υ2,Mq2)) can try to achieve unknown number a,
B, so as to obtain low frequency aberration MqThe specific functional relation between parameter υ, the low frequency aberration M inputted according to clientq
Value, by inverse function υ=f-1(Mq)=(Mq- b)/a can try to achieve the parameter υ general proportions of anchor point (i.e. needed for).
2) if δ-coordinates coordinates are quantified by 7bits or 8bits, now low frequency aberration MqWith the letter between parameter υ
Number Mq=f (v) exponentially forms：Mq=aebυ.It is similar can also to try to achieve a, b value by choosing two different parameter υ.Again
By inverse function υ=f-1(Mq)=(lnMq- lna)/b and client input MqValue, tries to achieve parameter υ.
3) if δ-coordinates coordinates are quantified by 6bits, for the model low frequency aberration M havingqBetween parameter υ
Function Mq=f (v) is linear, and some models then exponentially functional form.Therefore, we are fitted by the method for weighting
Curve now, i.e. υ=α υ1+(1-α)υ2, wherein：υ1=f-1(Mq)=(Mq- b)/a, υ2=f-1(Mq)=(lnMq-lna)/b
Step 5：To threedimensional model, actual low frequency aberration is finely adjusted.Due to the M' of real incomeqWith user's selection
Low frequency aberration MqIn the presence of certain deviation, low frequency aberration now is finely tuned using Greedy anchor points choosing method, until | | Mq-M'q
| |≤δ, wherein δ are gate valve value.The way of Greedy anchor point choosing methods is：Calculate M in reconstruction modelqIt is worth maximum summit,
And the summit is fixed as anchor point, reconstruction model selects M againqIt is worth maximum summit, the requirement until meeting client.
Claims (4)
1. the three-dimensional model transmitting method based on user's specification error precision, it is characterised in that：
Step 1, user is allowed to specify the collimation error and low frequency aberration of required mode in client；
Step 2, the geometric data of the model is done Laplacian conversions by service end, the coordinate after conversion be referred to as δ-
Coordinates coordinates；
Step 3, the collimation error selected according to user, corresponding quantization is done by δ-coordinates coordinates；
Step 4, by the anchor point of two different values of Selecting All Parameters υ, the function come between tectonic model low frequency aberration and parameter υ closes
System, and the low frequency aberration specified according to this function and user chooses the anchor point quantity added needed for the model, wherein parameter υ tables
Show anchor point dense degree；
Step 5, the low frequency aberration of the model is finely adjusted, it is met the requirement of user, and the model is sent into client
End；After client receives data, least square solution, the geometric data with regard to the model can be recovered are solved.
2. the three-dimensional model transmitting method according to claim 1 based on user's specification error precision, it is characterised in that：Step
Rapid 2 are specifically：
Define V={ v1,v2,,,vnBe grid M vertex set, wherein vi=(xi,yi,zi)；Define the δ on each summit-
Coordinates coordinates areWherein diThe number abutted directly against a little on summit is represented,
The as degree on summit,For xi,yi,ziCorresponding δ-coordinates coordinates；It is L=to define Laplacian matrixes
D-A, wherein D are main diagonal matrix, and the element on leading diagonal is：Dii=di, A is grid M adjacency matrix, and it is just like ShiShimonoseki
System：Lx=D δ(x), Ly=D δ(y), Lz=D δ(z)；Can be by the x of grid by three above relational expression, y, z coordinate be converted into δ-
Coordinates coordinates.
3. the three-dimensional model transmitting method according to claim 1 based on user's specification error precision, it is characterised in that：Step
Rapid 4 are specifically：
Tectonic model low frequency aberration MqFunctional relation M between parameter υq=f (v), it is quantified by δ-coordinates coordinates
The difference of precision is divided into three kinds of situations, and the low frequency aberration precision selected according to user obtains required anchor point quantity；
Wherein function Mq=f (v) three kinds of concrete conditions are as follows：
1) if δ-coordinates coordinates are quantified by 3bits, 4bits or 5bits, now low frequency aberration MqBetween parameter υ
Function Mq=f (v) is linear：Mq=av+b；Selecting All Parameters υ two different values, mould is reconstructed by solving least square method
Type, respectively obtain reconstruction model error amount Mq1,Mq2；By these two pair value ((υ1,Mq1), (υ2,Mq2)) unknown number a, b can be tried to achieve, from
And low frequency aberration M is obtainedqThe specific functional relation between parameter υ, the low frequency aberration M inputted according to clientqValue, by
Inverse function υ=f-1(Mq)=(Mq- b)/a can try to achieve parameter υ；
2) if δ-coordinates coordinates are quantified by 7bits or 8bits, now low frequency aberration MqFunction M between parameter υq
=f (v) exponentially forms：Mq=aebυ；By choosing two different parameter υ, a, b value are tried to achieve；Again by inverse function υ=f-1
(Mq)=(lnMq- lna)/b and client input MqValue, tries to achieve parameter υ；
3) if δ-coordinates coordinates are quantified by 6bits, for the model low frequency aberration M havingqFunction Mq between parameter υ
=f (v) is linear, and some models then exponentially functional form；Therefore, song now is fitted by the method for weighting
Line, i.e. υ=α υ1+(1-α)υ2, wherein：υ1=f-1(Mq)=(Mq- b)/a, υ2=f-1(Mq)=(lnMq-lna)/b。
4. the three-dimensional model transmitting method according to claim 1 based on user's specification error precision, it is characterised in that：Step
Rapid 5 finely tune low frequency aberration now using Greedy anchor points choosing method, until | | Mq-M'q| |≤δ, wherein δ are gate valve value；
Greedy anchor point choosing methods are：Calculate M in reconstruction modelqIt is worth maximum summit, and the summit is fixed as anchor point, weighs again
Structure model selects MqIt is worth maximum summit, the requirement until meeting client.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510057252.2A CN104715495B (en) | 2015-02-04 | 2015-02-04 | Three-dimensional model transmitting method based on user's specification error precision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510057252.2A CN104715495B (en) | 2015-02-04 | 2015-02-04 | Three-dimensional model transmitting method based on user's specification error precision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104715495A CN104715495A (en) | 2015-06-17 |
| CN104715495B true CN104715495B (en) | 2018-01-19 |
Family
ID=53414789
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510057252.2A Active CN104715495B (en) | 2015-02-04 | 2015-02-04 | Three-dimensional model transmitting method based on user's specification error precision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104715495B (en) |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101609563A (en) * | 2009-07-27 | 2009-12-23 | 浙江工商大学 | A kind of construction method of binary tree of 3 D model shape features |
| EP2924940B1 (en) * | 2012-12-27 | 2019-12-04 | Huawei Technologies Co., Ltd. | User plane data transmission methods, mobility management network element and evolved node b |
| CN103840871B (en) * | 2014-01-14 | 2017-01-11 | 浙江工商大学 | Encoding and reconstruction method of three-dimensional model topological data robust transmission in wireless network |
-
2015
- 2015-02-04 CN CN201510057252.2A patent/CN104715495B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN104715495A (en) | 2015-06-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108632621B (en) | A kind of point cloud genera compression method based on distinguishing hierarchy | |
| CN1684109B (en) | Method and apparatus for encoding and decoding 3d data | |
| CN108322742A (en) | A kind of point cloud genera compression method based on intra prediction | |
| CN101610411A (en) | A kind of method and system of video sequence mixed encoding and decoding | |
| CN113870437A (en) | Digital twin BIM model conversion and lightweight method and device | |
| CN104463954B (en) | A kind of analogy method and system of 3-D view surface details | |
| CN104012097A (en) | Method and apparatus for compressing and decompressing data | |
| CN108235057A (en) | A kind of network transmission and method for visualizing of extensive utilizing ocean current data | |
| CN105303616A (en) | Embossment modeling method based on single photograph | |
| CN106204735A (en) | Unity3D terrain data using method in Direct3D 11 environment | |
| CN109979008A (en) | A kind of point cloud band division methods based on attribute | |
| CN109540023A (en) | Object surface depth value measurement method based on two-value grid coding formwork structure light | |
| CN103927424A (en) | Method for simplifying building three-dimensional model structure by utilizing volume mesh | |
| CN108986218A (en) | A kind of building point off density cloud fast reconstructing method based on PMVS | |
| CN104899889A (en) | Video vectorization method based on tetrahedral mesh | |
| CN105160705B (en) | A kind of processing method based on d engine to mobile terminal 3 d image data | |
| CN104715495B (en) | Three-dimensional model transmitting method based on user's specification error precision | |
| CN107105245B (en) | High speed JPEG method for compressing image based on TMS320C6678 chip | |
| CN113392485A (en) | Three-dimensional virtual simulation method for underground water supply pipe network | |
| CN101795410A (en) | Texture compression and synthesis method with fine granularity and high compression rate | |
| CN104244008B (en) | Depth video coding method | |
| CN103761753B (en) | Decompression method based on texture image similarity | |
| WO2021115466A1 (en) | Point cloud data encoding method, point cloud data decoding method, storage medium and device | |
| CN113034627A (en) | Virtual structured light 3D point cloud compression method based on geometric rearrangement | |
| WO2023155778A1 (en) | Encoding method and apparatus, and device |
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 |