CN104658038B - 3-dimensional digital content intelligence production method and manufacturing system based on motion capture - Google Patents
3-dimensional digital content intelligence production method and manufacturing system based on motion capture Download PDFInfo
- Publication number
- CN104658038B CN104658038B CN201510108078.XA CN201510108078A CN104658038B CN 104658038 B CN104658038 B CN 104658038B CN 201510108078 A CN201510108078 A CN 201510108078A CN 104658038 B CN104658038 B CN 104658038B
- Authority
- CN
- China
- Prior art keywords
- motion capture
- action
- data
- matching
- actor model
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T13/00—Animation
- G06T13/20—3D [Three Dimensional] animation
- G06T13/40—3D [Three Dimensional] animation of characters, e.g. humans, animals or virtual beings
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/08—Indexing scheme for image data processing or generation, in general involving all processing steps from image acquisition to 3D model generation
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Processing Or Creating Images (AREA)
Abstract
A kind of 3-dimensional digital content intelligence production method based on motion capture disclosed by the invention, comprising the following steps: 1) 3D actor model pre-processes；2) data of motion capture are received, and carry out data intelligence processing；3) matching and pretreated 3D actor model needed for being loaded into, carry out Auto-matching and binding；4) driving 3D actor model is moved synchronously with performance person, and makees action correction；5) animation digital content exports.A kind of 3-dimensional digital content production system based on motion capture disclosed by the invention includes action data input module, animation actor model analysis module, movement matching module, action drives module, action data intelligent processing module, action correction module, action data output module.The quality that production method and manufacturing system of the invention can greatly improve three-dimensional cartoon producing efficiency, cost of manufacture is greatly reduced, greatly improves three-dimensional cartoon production, pushes the development of animation industry.
Description
Technical field
The invention belongs to 3-dimensional digital content intelligence manufacture technology field, especially a kind of three-dimensionals based on motion capture
Digital content intelligent production method and manufacturing system.
Background technique
Higher pursuit with the fast development and people of computer technology to quality of the life, more 3-dimensional digitals
Product has entered into people's daily life.In entertainment industry, high-tech animation video display and film trick production increasingly obtain people
Like pursued with market, the annual whole world reaches tens billion of dollars in the investment of three-dimensional animation industry.In movies-making system
In work, the shared ratio of 3-dimensional digital client computers production is increasing.Present animation video display are mostly motor-driven using calculating
Draw software, directly it is hand-made on computers come out.For a minister up to cartoon in 1 hour, then manual drawing is needed
Nearly 100,000 picture needs the investment of a large amount of manpower, time and fund, and since movement is manual drawing, is regarding
The feeling for still giving " animation " in feel lacks the true and naturality of figure action.Oscar winner in 2009 " Ah
It is all to reach " success show there is epoch-making meaning in motion picture production history, largely using advanced in " A Fanda " film
Three-dimensional motion capturing technology is accurately matched in cartoon role by the movement for performing true man and expression, has started animation
One new mileage of film making, fundamentally changes existing video display cartoon making or even special technology making method, increases
Be powered on shadow in figure action authenticity, increase the ornamental value of vision, to world spectators bring a completely new feeling and
It enjoys.Also huge business success is achieved, becomes first box-office value in world's film history and exceeds 2,000,000,000 dollars of pcmo films
Make.At present in the production of foreign films sheet, the digital content production based on motion capture is widely used.
3-dimensional digital content production based on motion capture mainly includes three key steps (i.e. motion capture, movement
With/binding, action drives) and some additional steps.Motion capture mainly captures system by three-dimensional motion and performs true man
Movement capture in real time；Movement matching needs that true man's body is matched and bound with cartoon role；Action drives
It is then that cartoon role is driven to make same movement by the movement of true man.Currently, the three-dimensional motion based on different technologies is caught
It catches system and has multiple product in market sale, such as Vicon, Giant, MotionAnalysis, Optitrack system etc.,
And technology is relatively mature.In movement matching and action drives, the Motionbuilder software of autodesk, inc. is
For carrying out the software tool of movement matching and action drives generally used.Producer is obtaining through motion capture system institute
After collecting figure action track data (.bvh format or .c3d format), Motionbuilder software is directly entered data into
Carry out manual action matching and action drives.
In Motionbuilder software in use, mainly there is the problem of following several respects:
At most of conditions, the body of performance people and the body of animation role are different, therefore
Motionbuilder needs manually to be matched the bone of performing artist and role's bone every time.
Motionbuilder needs the skeleton framework of the standard using autodesk, inc. oneself definition.But
In animation film, the skeletal architecture and skeleton framework of many anthropomorphic animal shapes are different.For these with
Skeleton framework difference animation role defined in Motionbuilder, Motionbuilder can not then carry out human action
Matching and action drives.
In movement matching and action drives, a common problem is movement " exposing the false " and " sliding steps ".Work as performance
Person is when carrying out the personal performance of trick, due to being physically usually very different for body shape and animation role,
When movement matching, the trick of role can be inserted into the certain positions of body and generate " exposing the false ", can also generate the step of animation role
It cannot correctly step on the ground and the phenomenon that " sliding steps " such as empty or drift about occur.
In many cases, when performing artist's body and larger animation actor model difference, it be easy to cause distortion and movement
Deformation, therefore the use of Motionbuilder software is being to need tie up by hand to performing artist's body and cartoon role model
It is fixed, time-consuming effort.System of the huge business success of " A Fanda " film from more than 400,000,000 dollar investments and up to four years
Make, wherein a big chunk time and fund are put into three-dimensional motion capture and cartoon making, especially in figure action
And in artificial matching/binding and action drives of expression and film appearances model.
Summary of the invention
The 3-dimensional digital content intelligence production method that the problem to be solved in the present invention is to provide a kind of based on motion capture and
Manufacturing system, the system and method can realize the automation of movement matching and action drives in 3-dimensional digital content production；And
It can be realized the customized of skeleton framework, so that body shape and animation role are more bonded, avoid " exposing the false " and " sliding
Step " phenomenon；Reach the efficiency and precision for improving 3-dimensional digital content production, is further fabricated to reaching reduction digital content
Originally, the purpose of time and raising production quality.
A kind of 3-dimensional digital method for manufacturing contents based on motion capture disclosed by the invention, comprising the following steps:
1) 3D actor model pre-processes, and obtains processed role's skeleton character information；These role's skeleton character information
To match and binding provides foundation.
2) it is loaded into and extracts the pretreated role's skeleton character information of 3D actor model, receive the data of motion capture；
3) data of the motion capture received are matched and is bound with 3D actor model；
4) driving 3D actor model is moved synchronously with performance person.
Method of the invention can realize Auto-matching and binding, and automatic driving.But it is complete to be not required for these movements
Portion is to be automatically performed, even if some processes are to manually complete, can also realize technical solution of the present invention.For example, matching and binding
Movement selection is automatically performed, and selection is driven to manually complete.
As a further improvement of the above technical scheme, the step 1) 3D actor model pretreatment includes: to 3D role
The surface geometry dough sheet of model is analyzed and processed, and extracts required 3D role's bone information, and calculate between bone node
Geometry dough sheet and the movement weight of accompanying bone node and the geometrical characteristic of bone, and by these processed 3D roles
Bone information is stored in the tree-shaped mechanism of bone；Reject the redundant information unrelated with motion capture.
As an alternative to the above-described technical solution, the step 3) matching and binding procedure include every by what is received
The skeleton motion information of one frame is matched and is bound with 3D actor model.
As an alternative to the above-described technical solution, the step 4) driving process includes preparatory according to 3D actor model
Geometry dough sheet and the movement weight of accompanying bone node and the geometrical characteristic of bone between calculated bone node,
The movement position of the space geometry of each geometry dough sheet of 3D actor model, driving 3D actor model and performance person are calculated in real time
It moves synchronously.
As an alternative to the above-described technical solution, method of the invention is further comprising the steps of:
5) while carrying out step 4) driving, character surface shape collision detection is carried out to matching result, is such as examined
Measuring body surface has collision part, then carries out collision Processing for removing.
As an alternative to the above-described technical solution, method of the invention is further comprising the steps of:
6) action correction: the movement position of 3D role's end artis is calculated, then passes through inverse kinematics (Inverse
Kinematics calculating) obtains the position of each artis.
As an alternative to the above-described technical solution, method of the invention is further comprising the steps of: 7) action data is defeated
Out.
As an alternative to the above-described technical solution, method of the invention is further comprising the steps of: in the step 1)
After the pretreatment of 3D actor model, before step 2 receives the data of motion capture, motion capture is carried out.
As an alternative to the above-described technical solution, method of the invention is further comprising the steps of:
Action data Intelligent treatment: after the data that the step 2 receives motion capture, for being hidden due to cursor point
Gear or other reasons and there is the part of loss of data, using the method for multi-frame analysis, utilize the before and after frames of lost data frames
Collected cursor point position, calculate the position for losing cursor point, fill omitted data；Jump for cursor point
It is dynamic, then data are smoothed automatically using signal smoothing algorithm, eliminate shake.
Manufacturing system for production method described in above-mentioned technical proposal, it is characterised in that: include:
Action data input and action data intelligent processing module are used to receive the data of motion capture, and to motion capture
Data carry out repair process；
Animation actor model intelligent analysis module, for analyzing the bone and body of animation role；
Matching module is acted, for matching and being tied in animation role by the human action captured；
Action drives module, driving 3D actor model are moved synchronously with performance person；
Action correction module is used for action correction；
Action data output module is exported for action data.
Compared with the existing 3-dimensional digital content production process based on motion capture, what the invention was proposed is caught based on movement
The specific biggish advantage of the 3-dimensional digital method for manufacturing contents and manufacturing system caught, is mainly manifested in:
Three-dimensional cartoon producing efficiency greatly improves, cost of manufacture is greatly reduced: making currently based on the animation of motion capture
It is main or carried out by animation teacher with softwares such as the MotionBuilder of Autodesk it is hand-made, it is driven catch data reparation,
The animations production processes such as action data roleization, movement matching and action correction, animation production per minute take around 10
The working time of hour.And the matching and drive system proposed through the invention, all animation production processes all can automatic intelligence
Energyization is completed, and animation production per minute need to only spend 5 minutes time, is improved efficiency hundred times or more, is also therefore substantially reduced
Human cost.In addition, greatly shortening animation fabrication cycle due to the raising of producing efficiency, huge market can also be brought to imitate
Fruit and economic benefit.
The quality of three-dimensional cartoon production greatly improves: since the animation production based on motion capture is all by several animation teachers
It has been manually done, each one technical level and experience of animation teacher will lead to whole animation, and to make quality ginseng time uneven, influences whole
The quality of body cartoon.Simultaneously as the technical level and experience of animation teacher, go to repair by hand, smoothly move and catch original number
According to and correction expose the false, the movement such as sliding steps, can all influence the quality of animation production to a certain extent.It is proposed by using the present invention
3-dimensional digital method for manufacturing contents based on motion capture and manufacturing system then can be significantly using the algorithm of system intelligent
Improve animation production quality.Simultaneously as all production are automatically performed by system, then animation production quality is greatly improved
Consistency.
Push the development of animation industry: most box-office hit in Oscar winner " A Fanda " acquisition histories in 2009
Afterwards, the animation production based on motion capture has been widely used in the production of foreign films sheet.In China, although from foreign countries
The high-end motion capture equipment of many sets has been introduced, but the animation video display of the production of the animation based on motion capture are not much, wherein
One most important reason is exactly still to have been manually done currently based on the animation manufacturing process of motion capture, with hand traditional at present
The method of work key frame is little compared to advantage.By using the 3-dimensional digital content system proposed by the invention based on motion capture
Make method and manufacturing system, is then greatly improved producing efficiency and quality, sufficiently shows the high-tech edge of motion capture, into one
Step promotes the application that this high-tech innovative technology of motion capture makes in digital animation, pushes the development of animation industry, it will
Generate great Social benefit and economic benefit.
Detailed description of the invention
Fig. 1 is that the present invention is based on the 3-dimensional digital method for manufacturing contents of motion capture and the block diagrams of system；
Fig. 2 is that the present invention is based on the 3-dimensional digital content production system structural schematic diagrams of motion capture.
Specific embodiment
With reference to the accompanying drawing and how with for the classical animation film " creating a tremendous uproar " of present invention production, the present invention is proposed
3-dimensional digital method for manufacturing contents based on motion capture and system be described in detail.
It paints from by traditional-handwork or different with animation software hand-made " creating an uproar in sky ", it, can be with this system
The performance of performer is acted and is captured by motion capture system, cartoon role " Sun Wukong " is given in Auto-matching and drives
" Sun Wukong " work synchronous with performer is corresponding to be acted.Following embodiment is merely illustrative, not to protection of the invention
Range causes to limit.
As illustrated in fig. 1 and 2, in matching and drive system hardware structure, three dimensions of the invention based on motion capture
Word content intelligence manufacturing system includes:
Intelligent Matching and drive system, three-dimensional (3D) motion capture system (dynamic to catch system) and storage system.
Matching and drive system are mainly the software systems of three-dimensional cartoon digital content a processing and production, be may operate at
In the computer equipment of one standard, comprising:
Action data input and action data intelligent processing module (MotionDataInputandMotion Data
ProcessingModue), for receiving the data of motion capture, and motion capture data is carried out to be automatically repaired processing.
Animation actor model analysis module (Character Model Analysis Module), for analyzing animation angle
The bone and body of color；
It acts matching module (Motion Matching Module), the human action Auto-matching for will be captured
Into animation role；
Action drives module (Motion Driven Module), for driving 3D actor model and performance person automatically
It moves synchronously；
Action correction module (Motion Tuning Module) is automatically corrected for acting；
Action data output module (Motion Data Output Module) is exported for action data.
As shown in Fig. 2, three-dimensional motion captures the motion capture system that system can be more mature using current industry technology,
The capture system includes multiple high speed high-definition cameras, the motion capture software placed around motion capture place, while
Multiple large display screen curtains are hung around motion capture place, are used to real-time display performing artist, and such as " Sun Wukong " action performer is dynamic
Make to capture effect and animation role " Sun Wukong " and match driving effect, in this way director, performing artist and moving catch technical staff can and
When adjust action effect, improve the efficiency and quality of animation production.Matching and drive system mainly pass through network data line and move
Make capture system and storage system carries out real-time data communication, can also be communicated by wireless network, to carry out real-time three
Tie up action data matching, action drives, movement processing and action correction.Storage system mainly includes pre-processing for saving
Animation actor model data and information database (server) and animation digital content/video/image for completing
Database (server).
Matching and drive system automatically analyze the bone and body of " Sun Wukong " animation role, the performing artist that will be captured
Human action is automatically matched in " Sun Wukong ", and synchronous driving " Sun Wukong " makees corresponding movement.
Matching and drive system also support customized animation role skeletal architecture, such as except standardized human body's skeletal architecture is supported
" the ox Prince of the Devils " model of " creating a tremendous uproar ", bone tree that can be open by this system, addition " the ox Prince of the Devils " it is small
The anti-bone of leg.Customized animation role skeletal architecture is completed by matching and drive module.Matching and drive system will be performing artist's bones
Corresponding skeleton point carries out right in each skeleton point and animation role " Sun Wukong " skeletal architecture of bone framework, and it is right that reservation fails
Skeleton point.
Matching and drive system have the function of real-time action matching and action drives.In performing artist's progress performance,
Movement can be matched into " Sun Wukong " in real time, role " Sun Wukong " is driven to make to move and be displayed on the screen accordingly in real time.This
Sample, the trick that performing artist according to the action effect of the animation role " Sun Wukong " of real-time display on screen, can adjust oneself are dynamic
Work is exposed the false to avoid movement.In addition, matching and drive system also will carry out character surface shape to matching result after movement matching
Shape collision detection (Collision Detection).For the body surface part detected, matching and drive system will be carried out
It collides Processing for removing (Collision Free).
Matching and drive system have the function of action data Intelligent treatment: for the cursor point with performance person by
In block and occur the phenomenon that loss of data or due to performance movement speed and body shake cause cursor point data obscure and
Situations such as bounce, matching and drive system can make automated intelligent processing to motion capture data: matching and drive system are using more
The method of frame analysis calculates the position for losing cursor point automatically, fills omitted data.In addition, matching and drive system are also adopted
Data are smoothed automatically with signal smoothing algorithm, eliminate shake.
Matching and drive system support the current various file output modes of industry, are then handled.For example, using common
Fbx file output mode, complete auto-action matching and action drives, matching and drive system are by " Sun Wukong " of generation
Action sequence is output in fbx file.This fbx file can be importing directly into the Motionbuilder of AutoDesk company
Secondary editor and processing are carried out, can also be importing directly into Maya or 3D Max and carry out editor and the processing of animation later rendering, with
Animation at present makes seamless connection.While the production in order to support two-dimensional anime, matching and drive system are also supported to show screen
Show that content is directly exported with two dimensional image format (such as .jpg .png etc.), or with video mode (such as .avi .mp4 etc.).
It should be known to those skilled in the art that also achievable of the invention using above-mentioned manufacturing system part components
Purpose, it is possible to use the software or hardware that the present embodiment does not describe complete the object of the invention.
In conjunction with attached drawing 1 and above-mentioned manufacturing system, to the 3-dimensional digital content proposed by the present invention based on motion capture
Production method is described in detail using exemplary embodiment.Step method cited by the present embodiment is not to the present invention
The limitation of the sequence and number of steps of method, the realization of the method for the present invention can carry out one or one in selection step as needed
Implement, also can adjust sequence therein.
3-dimensional digital method for manufacturing contents proposed by the present invention based on motion capture the following steps are included:
The pretreatment of 3D actor model: it captures carrying out real-time action with before content production, needs to the wanted matched angle 3D
Color model " Sun Wukong " pre-processes.Animation teacher with animation Software for producing (such as Maya or 3DMax etc.) complete animation and
It, can be by 3D role's " Sun Wukong " mode input to this system, animation role analysis module pair after 3D Role Modeling in game
The surface geometry dough sheet of " Sun Wukong " model is analyzed and processed, and extracts required angle by calculating the curvature value of each dough sheet
Color skeleton character information, and according to two joint endpoints of geometry dough sheet and place bone at a distance from, pass through thermodynamics diffusion equation
The movement weight (between 0 to 1) and bone of the automatic geometry dough sheet calculated between bone node and accompanying bone node
Geometrical characteristic (for example, size, position, axial direction etc.), and by these information preservations in the tree-shaped mechanism of bone, it is real-time to increase
The matching speed and accuracy of motion capture.Meanwhile rejecting redundant information (such as model material, background unrelated with motion capture
Light, camera position etc.) to improve matching speed, reduce memory space.Final matching and drive system output are customized
Meet to move in real time and catch required 3D " Sun Wukong " actor model, and model is stored in 3D actor model database in case in real time
It is used when motion capture.
Real-time action is captured, matches and driven: matching and drive system of the invention can be widely used with current industry
Real-time action capture system (the Motion Analysis and Optitrack in such as U.S. are dynamic to catch system) docked.For
The quality and speed of motion capture are improved, motion capture system and matching drive system are separately operable on a different computer
(it is also feasible for certainly, operating on the same computer), carries out data exchange and transmission by wired or wireless network.When
When carrying out real-time action capture and content production, matching and pretreated 3D needed for matching drive system is first loaded into
Then role " Sun Wukong " model carries out communication connection with dynamic system of catching.When carrying out motion capture, motion capture system passes through
Included real-time Data Transmission plug-in unit is by the motion information real-time transmission of the bone of performing artist and accompanying cursor point to system
Action data input and intelligent processing module.After matching and drive system receive the data of motion capture, matching module is acted
By the pretreated skeleton character information of extraction " Sun Wukong " 3D actor model, the skeleton motion of each frame received is believed
Breath carries out Auto-matching and binding with " Sun Wukong " 3D actor model: for each bone of performing artist and joint (such as hand, foot,
Trunk, head etc.), the corresponding bone of " Sun Wukong " model and pass are found out according to position of the bone on bone tree
Section, is allowed to match and bind one by one.After completing matching and binding, action drives module will be counted in advance according to 3D actor model
Geometry dough sheet and the movement weight of accompanying bone node and the geometrical characteristic of bone between the bone node of calculating, meter
The movement position of the space geometry of each geometry dough sheet of 3D actor model is calculated, driving " Sun Wukong " model and performance person make same
Step movement.Meanwhile the intelligent driver module of matching and drive system will also carry out the collision inspection of character surface shape to matching result
Survey (Collision Detection), judgement " Sun Wukong " model during the motion, each geometry dough sheet whether and mould
Other geometry dough sheets of type intersect.For the body surface part detected, system will carry out collision Processing for removing
(Collision Free).If two geometry dough sheet intersections, outwardly move one of dough sheet by its normal direction
Dynamic, until two dough sheets no longer intersect, realization to expand elimination significantly.
Action data Intelligent treatment: movement range and angle and locating camera site due to performance person, movement
Cursor point with performing artist the phenomenon that loss of data may can not occur by cameras capture due to blocking.In addition, by
In the performance movement speed of performing artist and the shake of body, it is likely to result in obscuring and beating for cursor point data.It is obtained in real time
It must move after catching data, matching and drive system will make automated intelligent post-processing to motion capture data.For being hidden due to cursor point
It keeps off and the part of loss of data occurs, will be collected using the method for multi-frame analysis using the before and after frames of lost data frames
Cursor point position, it is automatic to calculate the position for losing cursor point, fill omitted data.Bounce for cursor point, then
Data are smoothed automatically using signal smoothing algorithm, eliminate shake.
Action correction: in general by the animation of the 3 dimensional coil geometry of performance person and the driving of being matched
The 3 dimensional coil geometry of role is different from size and ratio, movement bound in cartoon role will appear expose the false, sliding steps
With incorrect operating position.Matching and drive system according to the geometrical characteristic information of every bone of the performing artist advanced for,
The three-dimensional geometry position of the geometrical characteristic information of every bone of cartoon role, the joint action of performing artist, it is automatic to calculate animation angle
The movement position of the end artis such as color hand, foot, then pass through the calculating of inverse kinematics (Inverse Kinematics), it obtains every
The position of a artis.Cartoon role binding and driving institute will be efficiently solved by the action correction of matching and drive system
The incorrect movement problems such as bring exposes the false, sliding steps.Real time correction can be carried out, in action drives with the accurate of safety action
Property.
Action data output: matching and drive system support the current common file output mode of industry, are then located
Reason.Such as fbx file output mode, after completing auto-action matching and action drives, matching and drive system are by generation
Role action sequence is output in fbx file.This fbx file can be importing directly into AutoDesk company
Secondary editor and processing are carried out in Motionbuilder, can also be importing directly into Maya or 3D Max and be carried out editor and animation
Later rendering processing makes seamless connection with current animation.While the production in order to support two-dimensional anime, matching and drive system
Also support by screen display content with two dimensional image format (for example .jpg .png etc.) or with video mode (for example .avi,
.mp4 etc.) directly export.
Three-dimensional (3D) motion capture system (dynamic to catch system) is that (multiple high-speed cameras, movement are caught by motion capture equipment
Catch software and motion capture real-time display equipment (such as large screen LCD display).Current foreign countries Vicon, Giant,
The companies such as MotionAnalysis, Optitrack can provide motion capture equipment and corresponding motion capture software.Pass through
Cable network or wireless network, motion capture system are attached with proposed by the present invention match with drive system, motion capture
Data afterwards can be driven and be matched in real time with animation role, made digital content can also real-time display caught in movement
On the giant display at the scene of catching, to instruct the actual effect after performer or the production of producer's real-time experience number, help real
When corrective action perform, improve animation digital content production efficiency and quality.
Storage system: it is mainly made of animation character data library and animation digital content data library.Animation character data library
It is to be matched for storing pretreated three-dimensional cartoon role skeleton character information for capturing data with real-time action.It is dynamic
Unrestrained digital content data library can be such as directly loaded into for storing the data content completed through making steps such as overmatching drivings
Perhaps action video, image etc. in the fbx 3-dimensional digital of later rendering are carried out to Maya.
There are many concrete application approach of the present invention, the above is only a preferred embodiment of the present invention, it is noted that for
For those skilled in the art, without departing from the principle of the present invention, several improvement can also be made, this
A little improve also should be regarded as protection scope of the present invention.
Claims (4)
1. a kind of 3-dimensional digital content intelligence production method based on motion capture, it is characterised in that: the following steps are included:
1) 3D actor model pre-processes, and obtains processed role's skeleton character information；
2) it is loaded into and extracts the pretreated role's skeleton character information of 3D actor model, receive the data of motion capture；
3) data of the motion capture received are matched and is bound with 3D actor model；
4) driving 3D actor model is moved synchronously with performance person；
5) action correction: geometrical characteristic information according to every bone of the performing artist advanced for of matching and drive system, animation
The three-dimensional geometry position of the geometrical characteristic information of every bone of role, the joint action of performing artist calculates 3D role's end artis
Movement position, then by the calculating of inverse kinematics Inverse Kinematics, obtain the position of each artis；
The step 1) 3D actor model pretreatment includes: to be analyzed and processed to the surface geometry dough sheet of 3D actor model, is mentioned
Required 3D role's bone information is taken, and calculates the movement weight of the geometry dough sheet between bone node Yu accompanying bone node
And the geometrical characteristic of bone, and these processed 3D role's bone informations are stored in the tree-shaped mechanism of bone；It rejects
For the redundant information unrelated with motion capture after step 1) 3D actor model pretreatment, step 2 receives the number of motion capture
According to before, motion capture is carried out；
The step 2 receive motion capture data after, for due to cursor point block or other reasons and there are data
The part of loss, using the method for multi-frame analysis, using lost data frames before and after frames collected cursor point position,
The position for losing cursor point is calculated, omitted data are filled；Bounce for cursor point, then it is automatic using signal smoothing algorithm
Data are smoothed, shake is eliminated；
Step 3) matching and binding procedure include each frame that will be received skeleton motion information and 3D actor model into
Row matching and binding；
Step 4) driving process include geometry dough sheet between the bone node precomputed according to 3D actor model with
The movement weight of accompanying bone node and the geometrical characteristic of bone calculate each geometry dough sheet of 3D actor model in real time
The movement position of space geometry, driving 3D actor model are moved synchronously with performance person.
2. the 3-dimensional digital content intelligence production method according to claim 1 based on motion capture, it is characterised in that: also
The following steps are included:
While carrying out step 4) driving, character surface shape collision detection is carried out to matching result, such as detects shape
There is collision part in body surface face, then carries out collision Processing for removing.
3. the 3-dimensional digital content intelligence production method according to claim 1 based on motion capture, it is characterised in that: also
The following steps are included:
Action data output.
4. the manufacturing system for production method described in claims 1 to 3 any one, it is characterised in that: include:
Action data input and intelligent processing module, are repaired for receiving the data of motion capture, and to motion capture data
Multiple processing；
Animation actor model analysis module, for analyzing the bone and body of animation role；
Matching module is acted, for the human action captured to be matched in animation role；
Action drives module, driving 3D actor model are moved synchronously with performance person；
Action correction module is used for action correction；
Action data output module is exported for action data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510108078.XA CN104658038B (en) | 2015-03-12 | 2015-03-12 | 3-dimensional digital content intelligence production method and manufacturing system based on motion capture |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510108078.XA CN104658038B (en) | 2015-03-12 | 2015-03-12 | 3-dimensional digital content intelligence production method and manufacturing system based on motion capture |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104658038A CN104658038A (en) | 2015-05-27 |
| CN104658038B true CN104658038B (en) | 2019-01-18 |
Family
ID=53249110
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510108078.XA Active CN104658038B (en) | 2015-03-12 | 2015-03-12 | 3-dimensional digital content intelligence production method and manufacturing system based on motion capture |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104658038B (en) |
Families Citing this family (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105931283B (en) * | 2016-04-22 | 2019-10-29 | 南京梦宇三维技术有限公司 | A kind of 3-dimensional digital content intelligence production cloud platform based on motion capture big data |
| CN106200983B (en) * | 2016-07-25 | 2019-06-18 | 圣域科技（深圳）有限公司 | A kind of system of combination virtual reality and BIM realization virtual reality scenario architectural design |
| CN106530377B (en) * | 2016-12-16 | 2020-02-07 | 厦门黑镜科技有限公司 | Method and apparatus for manipulating three-dimensional animated characters |
| CN106799048A (en) * | 2017-03-30 | 2017-06-06 | 苏州乐米信息科技股份有限公司 | 3D action generation method and system based on dynamic acquisition |
| CN107194979A (en) * | 2017-05-11 | 2017-09-22 | 上海微漫网络科技有限公司 | The Scene Composition methods and system of a kind of virtual role |
| CN107274464A (en) * | 2017-05-31 | 2017-10-20 | 珠海金山网络游戏科技有限公司 | A kind of methods, devices and systems of real-time, interactive 3D animations |
| CN107248195A (en) * | 2017-05-31 | 2017-10-13 | 珠海金山网络游戏科技有限公司 | A kind of main broadcaster methods, devices and systems of augmented reality |
| CN107197385A (en) * | 2017-05-31 | 2017-09-22 | 珠海金山网络游戏科技有限公司 | A kind of real-time virtual idol live broadcasting method and system |
| CN107248185A (en) * | 2017-05-31 | 2017-10-13 | 珠海金山网络游戏科技有限公司 | A kind of virtual emulation idol real-time live broadcast method and system |
| CN107277599A (en) * | 2017-05-31 | 2017-10-20 | 珠海金山网络游戏科技有限公司 | A kind of live broadcasting method of virtual reality, device and system |
| CN107274465A (en) * | 2017-05-31 | 2017-10-20 | 珠海金山网络游戏科技有限公司 | A kind of main broadcaster methods, devices and systems of virtual reality |
| CN107274466A (en) * | 2017-05-31 | 2017-10-20 | 珠海金山网络游戏科技有限公司 | The methods, devices and systems that a kind of real-time double is caught |
| CN107527542B (en) * | 2017-09-18 | 2020-09-08 | 南京梦宇三维技术有限公司 | Percussion training system based on motion capture |
| CN108038160B (en) * | 2017-12-06 | 2019-03-08 | 央视动画有限公司 | Dynamic animation store method, dynamic animation call method and device |
| CN108288300A (en) * | 2018-01-12 | 2018-07-17 | 北京蜜枝科技有限公司 | Human action captures and skeleton data mapped system and its method |
| CN108198234B (en) * | 2018-01-12 | 2021-06-25 | 北京蜜枝科技有限公司 | Virtual character generating system and method capable of realizing real-time interaction |
| CN108876815B (en) * | 2018-04-28 | 2021-03-30 | 深圳市瑞立视多媒体科技有限公司 | Skeleton posture calculation method, character virtual model driving method and storage medium |
| CN108765530A (en) * | 2018-05-24 | 2018-11-06 | 北京声影动漫科技有限公司 | A method of making caricature and/or 2 D animation |
| CN108986190A (en) * | 2018-06-21 | 2018-12-11 | 珠海金山网络游戏科技有限公司 | A kind of method and system of the virtual newscaster based on human-like persona non-in three-dimensional animation |
| CN112866286B (en) * | 2018-10-29 | 2023-03-14 | 深圳市瑞立视多媒体科技有限公司 | Data transmission method and device, terminal equipment and storage medium |
| CN110033505A (en) * | 2019-04-16 | 2019-07-19 | 西安电子科技大学 | A kind of human action capture based on deep learning and virtual animation producing method |
| CN110473276B (en) * | 2019-06-28 | 2023-09-26 | 中影年年(北京)文化传媒有限公司 | High-efficiency three-dimensional cartoon production method |
| CN111243067A (en) * | 2020-01-10 | 2020-06-05 | 安徽五峰杰传媒科技有限公司 | High-immersion interactive action content production method |
| CN112862935B (en) * | 2021-03-16 | 2023-03-17 | 天津亚克互动科技有限公司 | Game role movement processing method and device, storage medium and computer equipment |
| CN113490054B (en) * | 2021-07-01 | 2024-07-09 | 网易（杭州）网络有限公司 | Virtual character control method, device, equipment and storage medium |
| CN114782284B (en) * | 2022-06-17 | 2022-09-23 | 广州三七极耀网络科技有限公司 | Motion data correction method, device, equipment and storage medium |
| CN115100331A (en) * | 2022-06-28 | 2022-09-23 | 功夫动漫股份有限公司 | Method and device for manufacturing personalized three-dimensional animation |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104021584B (en) * | 2014-06-25 | 2017-06-06 | 无锡梵天信息技术股份有限公司 | A kind of implementation method of Skeletal Skinned Animation |
-
2015
- 2015-03-12 CN CN201510108078.XA patent/CN104658038B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN104658038A (en) | 2015-05-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104658038B (en) | 3-dimensional digital content intelligence production method and manufacturing system based on motion capture | |
| CN110458939B (en) | Indoor scene modeling method based on visual angle generation | |
| CN108320297B (en) | Video target real-time tracking method and system | |
| CN110428493B (en) | Single-image human body three-dimensional reconstruction method and system based on grid deformation | |
| CN105931283B (en) | A kind of 3-dimensional digital content intelligence production cloud platform based on motion capture big data | |
| CN110555434B (en) | Method for detecting visual saliency of three-dimensional image through local contrast and global guidance | |
| CN109919887B (en) | Unsupervised image fusion method based on deep learning | |
| Livny et al. | Automatic reconstruction of tree skeletal structures from point clouds | |
| Kelly et al. | FrankenGAN: guided detail synthesis for building mass-models using style-synchonized GANs | |
| CN106997605B (en) | A method of foot type video is acquired by smart phone and sensing data obtains three-dimensional foot type | |
| CN107067429A (en) | Video editing system and method that face three-dimensional reconstruction and face based on deep learning are replaced | |
| CN109191369A (en) | 2D pictures turn method, storage medium and the device of 3D model | |
| US20170053438A1 (en) | Method and system for reconstructing a three-dimensional model of point clouds | |
| CN110728670A (en) | Low-voltage equipment operation and maintenance method based on AR technology | |
| US20070132765A1 (en) | Method of representing and animating two-dimensional humanoid character in three-dimensional space | |
| CN109800770A (en) | A kind of method, system and device of real-time target detection | |
| CN108830170B (en) | End-to-end target tracking method based on layered feature representation | |
| CN103646421A (en) | Tree lightweight 3D reconstruction method based on enhanced PyrLK optical flow method | |
| CN112528379A (en) | Building model generation method and device and electronic equipment | |
| CN115082254A (en) | Lean control digital twin system of transformer substation | |
| WO2024148732A1 (en) | Reconstruction method and apparatus for relightable implicit human body model | |
| CN114758070B (en) | Cross-domain multitasking-based single image three-dimensional human body fine reconstruction method | |
| CN115496862A (en) | Real-time three-dimensional reconstruction method and system based on SPIN model | |
| CN110363833A (en) | A kind of complete human body sport parameter representation method based on local rarefaction representation | |
| CN105740450A (en) | Multi-Kinect based 3D human body posture database construction method |
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 |