CN115358621A - Crowd formation planning method and system based on characteristic anchor points - Google Patents
Crowd formation planning method and system based on characteristic anchor points Download PDFInfo
- Publication number
- CN115358621A CN115358621A CN202211073984.7A CN202211073984A CN115358621A CN 115358621 A CN115358621 A CN 115358621A CN 202211073984 A CN202211073984 A CN 202211073984A CN 115358621 A CN115358621 A CN 115358621A
- Authority
- CN
- China
- Prior art keywords
- crowd
- edge
- loss
- anchor
- mapping
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06313—Resource planning in a project environment
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/20—Drawing from basic elements, e.g. lines or circles
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Marketing (AREA)
- Biodiversity & Conservation Biology (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a crowd formation planning method based on characteristic anchor points, which comprises the following steps: generating N frames of patterns corresponding to the crowd arrangement shape according to the time sequence; determining the point location distribution of the crowd corresponding to each frame of pattern; determining a mapping scheme from the ith frame to the (i + 1) th frame of pattern crowd point positions, specifically comprising selecting a characteristic anchor point set, determining a mapping relation of the anchor point set, defining a loss function of crowd formation transformation according to anchor points, and obtaining a macroscopic planning result with the lowest total loss; and obtaining the moving direction and speed of each person according to the mapping scheme. The invention provides a population formation planning method and system based on a characteristic anchor point, and provides an unbalanced population formation planning method and system with high controllability.
Description
Technical Field
The invention relates to a crowd formation planning method and system, in particular to a crowd formation planning method and system based on feature anchor points.
Background
In recent years, crowd formation transformation and planning have been developed in the fields of movie and television special effects, large-scale artistic performance (such as Olympic games on scenes), unmanned aerial vehicle formation performance and the like. Due to the introduction of space-time constraint of group motion, the group formation transformation has the characteristics of low motion loss, collision avoidance, shape adaptation and the like. Thus, crowd movement based on crowd formation transformation can often present a higher aesthetic visual perception than free crowd movement.
The basic idea of the existing research of formation transformation and planning of people groups is to specify a series of key processes, and people can be transformed into target shapes from initial shapes through smooth and collision-free/less motion transformation by a certain transformation method. A great deal of work is done to solve this process by bipartite matching optimization or constraint-based dynamic collaborative optimization methods. However, the existing methods have two major drawbacks: 1) The accessibility of the transformation process is ensured only by optimizing the overall consumption of the movement or keeping the pattern boundary, so that the integrity and continuity of people in the movement process can be kept between simple and similar pattern transformations, but the effect between the patterns with complex edges and low shape similarity is poor; 2) The group formation transformation process generated by the existing method is lack of flexibility, and the group formation transformation process with diversity cannot be generated for the given pattern constraint. This is a significant limitation to the user's design experience.
In fact, team behavior based on formation transformation has been widely applied to movie and television art creation and crowd performance. The existing related researches generate fixed group formation transformation results through a given pipeline, and the process is rigid and lacks creativity. For example, a method based on PLE criterion uses the overall motion loss of the group as a core optimization target for obtaining the formation transformation result, and actually, increasing a certain overall motion loss of the group may generate a more characteristic and aesthetic group motion. In fact, the evaluation index of the formation transformation of the group of people in the prior work is too single, the aesthetic feeling of the group movement is difficult to measure, and the generation of the result of diversity is a very effective means for obtaining the 'best' result.
Disclosure of Invention
The invention aims to provide a non-uniform and high-controllability crowd formation planning method and system aiming at the defects of the prior art, and the method and the system can generate high-diversity crowd motions based on established constraints, namely realize differentiated crowd formation transformation process planning.
The invention provides a crowd formation planning method based on characteristic anchor points, which comprises the following steps:
step one, generating N frames of patterns corresponding to the crowd arrangement shape according to a time sequence;
secondly, determining the point location distribution of the crowd corresponding to each frame of pattern;
step three, determining a mapping scheme from the ith frame to the (i + 1) th frame pattern crowd point location, wherein the specific method comprises the following steps:
3.2, determining the mapping relation of anchor point sets in the two frames of patterns;
3.3, defining a loss function of the crowd formation transformation according to the anchor point:
for the population F i Each non-anchor individual agent of ∈ A' i Then population F i Position p in i To the crowd F i+1 Middle position p i+1 Distance loss D of loss And azimuth loss O loss is fixed The meaning is:
defining a loss function:
V loss =εD loss +λO loss +μD
wherein epsilon, lambda and mu are proportional parameters, D is the measurement space distance from agt to agt' in the traditional method,
3.4, obtaining F i All agent pairs F in i+1 V of all positions loss, Obtaining a macroscopic planning result with the lowest total loss;
and step four, obtaining the movement direction and speed of each person according to the mapping scheme.
According to the population formation planning method provided by the embodiment of the invention, the method for selecting the characteristic anchor point set A in the step 3.1 comprises the following steps:
setting the number k of target anchor points to the crowd F i And F i+1 Dividing two groups of people into k subgroups respectively through k-means clustering, and then separating the distance from the gravity center of each subgroupThe nearest individual serves as an anchor point; center of gravity of subgroupThe components in each coordinate direction are:
X i is the position component in the X direction of any individual in the subgroup, Y i For the position component in the Y direction, and so on, W i Calculating weight parameters of different individuals when calculating the gravity center;
the method for determining the anchor point set mapping relation in the two-frame pattern in the step 3.2 comprises the following steps:
anchor point set A is obtained through calculation i And A i+1 And then, taking the negative value of the distance between the individual world coordinate points in the two sets as a weight to establish a complete bipartite graph, and obtaining the optimal matching result of the total weight as the mapping relation between the anchor points.
According to the population formation planning method provided by the embodiment of the invention, the method for selecting the characteristic anchor point set A and determining the mapping relation of the anchor point set in the steps 3.1 and 3.2 comprises the following steps:
acquiring the Edge of a crowd; for the crowd F i And F i+1 Edge of (2) i And Edge i+1 Respectively carrying out down-sampling to ensure Edge after sampling' i And Edge' i+1 The equipotential is satisfied:
Edge′ i ~Edge′ i+1
since crowd edges are continuous, edge' i And Edge' i+1 Can be regarded as two circular queues with equal length, and a certain agent is specified to map: edge' i (j)→Edge′ i+1 (k) And the other agents complete the mapping according to the sequence in the queue.
According to the population formation planning method provided by the embodiment of the invention, the optimal matching result of the total weight is obtained through a Kuhn-Munkres algorithm and is used as the mapping relation between anchor points.
The invention also provides a crowd formation planning system based on the characteristic anchor point, which comprises the following steps:
the pattern generating module is used for generating N frames of patterns corresponding to the crowd arrangement shape according to the time sequence;
the point location setting module is used for determining the point location distribution of the crowd corresponding to each frame of pattern;
the point location mapping module is used for determining a mapping scheme from the ith frame to the (i + 1) th frame pattern crowd point location;
the motion conversion module is used for obtaining the motion direction and the motion speed of each person according to the mapping scheme;
wherein, the point location mapping module includes:
anchor selection means for selecting a set of characteristic anchors among a population FThe non-anchor set is A' = F-A;
the anchor point mapping component is used for determining the mapping relation of the anchor point set in the two-frame pattern;
a loss function component for defining a loss function of the population formation transform according to the anchor point:
for the population F i Each non-anchor individual agent of ∈ A' i Then population F i Position p in i To the crowd F i+1 Middle position p i+1 Distance loss D of loss And azimuth loss O loss Is defined as follows:
defining a loss function:
V loss =εD loss +λO loss +μD
wherein epsilon, lambda and mu are proportional parameters, D is the measurement space distance from agt to agt' in the traditional method,
population planning means for obtaining F i All agent pairs F i+1 V of all positions loss And obtaining a macroscopic planning result with the lowest total loss.
According to the population formation planning system provided by the embodiment of the invention, the method for selecting the characteristic anchor point set A by the anchor point selection component comprises the following steps:
setting the number k of target anchor points to a crowd F i And F i+1 Dividing two groups of people into k subgroups by k-means clustering respectively, and then separating the distance from the gravity center of each subgroupThe nearest individual serves as an anchor point; center of gravity of subgroupThe components in each coordinate direction are:
X i is the position component in the X direction of any individual in the subgroup, Y i For the position component in the Y direction, and so on, W i Calculating weight parameters of different individuals when calculating the gravity center;
the anchor mapping component calculates an anchor set A i And A i+1 And then, taking the negative value of the distance between the individual world coordinate points in the two sets as a weight to establish a complete bipartite graph, and obtaining the optimal matching result of the total weight as the mapping relation between anchor points.
According to the crowd formation planning system provided by the embodiment of the invention, the anchor point selection component acquires the Edge of the crowd and performs the crowd F i And F i+1 Edge of (2) i And Edge i+1 Respectively carrying out down-sampling to ensure Edge after sampling' i And Edge' i+1 The equipotential is satisfied:
Edge′ i ~Edge′ i+1
will Edge' i And Edge' i+1 As a population F i And F i+1 The set of anchor points of (a);
edge 'since the crowd Edge is continuous' i And Edge' i+1 Can be regarded as two circular queues with equal length, the anchor mapping component designates one of the agents to map: edge' i (j)→Edge′ i+1 (k) And the other agents complete mapping according to the sequence in the queue.
According to the crowd formation planning system provided by the embodiment of the invention, the anchor point mapping component obtains the optimal matching result of the total weight as the mapping relation between anchor points through a Kuhn-Munkres algorithm.
The invention further provides a terminal, which comprises a processor, an input device, an output device and a memory, wherein the memory is used for storing program codes, and the processor is configured to run the program codes and execute the people group formation planning method.
The present invention also provides a computer-readable storage medium having stored thereon program instructions which, when executed by a processor, cause the processor to perform the aforementioned method of crowd formation planning.
Advantageous effects
The invention provides a population formation planning method and system based on a characteristic anchor point, and provides an unbalanced population formation planning method and system with high controllability.
Drawings
FIG. 1 is a flow chart of a population formation planning method based on feature anchor points;
FIG. 2 is a schematic diagram of a point location distribution scheme obtained from a pattern;
FIG. 3 is a flowchart of a method for determining a people point location mapping scheme for an ith frame to an (i + 1) th frame pattern;
FIG. 4 is a schematic diagram of a matching map of an edge selection method.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In order to obtain the effect of transforming the formation of the crowd with strong diversity, high boundary retentivity and low overall motion loss, embodiment 1 of the present invention provides a method for planning the formation of the crowd based on a feature anchor point, as shown in fig. 1, including the following steps:
generating N frames of patterns corresponding to the crowd arrangement shape according to a time sequence;
secondly, determining the point location distribution of the crowd corresponding to each frame of pattern; as shown in fig. 2, different point location distribution schemes can be obtained according to different population densities according to a pattern.
Step three, determining a mapping scheme from the ith frame to the (i + 1) th frame pattern crowd point, as shown in fig. 3, the specific method is as follows:
3.2, determining the mapping relation of anchor point sets in the two frames of patterns;
3.3, defining a loss function of the crowd formation transformation according to the anchor point:
after the crowd characteristic anchor point is selected and the mapping relation is determined, the whole crowd formation transformation can be calculated according to the anchor point. Since anchor points often account for only a very small portion (5% or even less) of the population as a whole, this process can be considered as a generalization of the small sample feature planning result to the population as a whole.
For the population F i Each non-anchor individual agent in (a) is belonged toA′ i Then the population F i Position p in i To the crowd F i+1 Middle position p i+1 Distance loss D of loss And the azimuth loss O loss Is defined as follows:
defining a loss function:
V loss =εD loss +λO loss +μD
wherein epsilon, lambda and mu are proportional parameters, D is the measurement space distance from agt to agt' in the traditional method,
V loss can measure F i Any of the agents in anchor set A i Under the influence of (2) at F i+1 Any position in the table. V loss The lower agent has a greater tendency to move to that location in the formation transformation. V loss The influence of the anchor point is introduced on the basis of distance D measurement in the traditional method, so that the algorithm has a control degree far higher than that of the traditional method in the process of the formation transformation of the human team.
3.4 obtaining F i All agent pairs F i+1 V of all positions loss Obtaining a macroscopic planning result with the lowest total loss;
and step four, obtaining the movement direction and speed of each person according to the mapping scheme.
After the point location mapping scheme of every two adjacent frame patterns is obtained, the motion direction and speed of each person can be obtained according to the position of each point at different time, and therefore effective crowd formation planning is conducted.
The embodiment 2 of the invention provides a crowd formation planning method, and the method for selecting the characteristic anchor point set A in the step 3.1 comprises the following steps:
setting the number k of target anchor points to the crowd F i And F i+1 Dividing two groups of people into k subgroups by k-means clustering respectively, and then separating the distance from the gravity center of each subgroupThe nearest individual serves as an anchor point; center of gravity of subgroupThe components in each coordinate direction are:
X i is the position component in the X direction of any individual in the subgroup, Y i For the position component in the Y direction, and so on, W i Weight parameters for different individuals when calculating the center of gravity;
the method for determining the anchor point set mapping relationship in the two frame patterns in step 3.2 comprises the following steps:
anchor point set A is obtained through calculation i And A i+1 And then, taking the negative value of the distance between the individual world coordinate points in the two sets as a weight to establish a complete bipartite graph, and obtaining the optimal matching result of the total weight as the mapping relation between the anchor points.
The method provided in embodiment 2 is suitable for convex polygons with simple edges, or is used in combination with other methods as a method for selecting feature points in a group.
The embodiment 3 of the invention provides a population formation planning method different from that of the embodiment 2, and the method for selecting the characteristic anchor point set A and determining the mapping relation of the anchor point set in the steps 3.1 and 3.2 comprises the following steps:
acquiring the Edge of a crowd; for the crowd F i And F i+1 Edge of (2) i And Edge i+1 Respectively down-sampling to ensure Edge after sampling' i And Edge' i+1 The equipotential is satisfied:
Edge′ i ~Edge′ i+1
since crowd edges are continuous, edge' i And Edge' i+1 Can be regarded as two circular queues with equal length, and a certain agent is specified to map: edge' i (j)→Edge′ i+1 (k) And the other agents complete the mapping according to the sequence in the queue. The matching map of the edge selection method in example 3 is shown in fig. 4.
According to the crowd formation planning method in the embodiment 4 of the invention, the optimal matching result of the total weight is obtained through a Kuhn-Munkres algorithm and is used as the mapping relation between anchor points.
The invention provides a non-equilibrium and high-controllability group formation transformation method based on the characteristics of people. Through different crowd characteristic selection schemes, a crowd formation transformation process can be designed according to overall maintenance, edge maintenance or other different requirements. Compared with the existing scheme, the method can realize more accurate group motion control, and can generate various interesting results aiming at specific pattern design.
The invention also provides a crowd formation planning system based on the characteristic anchor point, which comprises the following steps:
the pattern generation module is used for generating N frames of patterns corresponding to the crowd arrangement shape according to the time sequence;
the point location setting module is used for determining the point location distribution of the crowd corresponding to each frame of pattern;
the point location mapping module is used for determining a mapping scheme from the ith frame to the (i + 1) th frame pattern crowd point location;
the motion conversion module is used for obtaining the motion direction and the motion speed of each person according to the mapping scheme;
wherein, the point location mapping module includes:
anchor selection means for selecting a set of characteristic anchors in the population FThe non-anchor set is A' = F-A;
the anchor point mapping component is used for determining the mapping relation of the anchor point set in the two frames of patterns;
a loss function component for defining a loss function of the population formation transform according to the anchor point:
for the population F i Each non-anchor individual agent of ∈ A' i Then population F i Position p in i To the crowd F i+1 Middle position p i+1 Distance loss D of loss And the azimuth loss O loss Is defined as:
defining a loss function:
V loss =εD loss +λO loss +μD
wherein epsilon, lambda and mu are proportional parameters, D is the measurement space distance from agt to agt' in the traditional method,
population planning means for obtaining F i All agent pairs F i+1 V of all positions loss And obtaining a macroscopic planning result with the lowest total loss.
According to the crowd formation planning system provided by the embodiment of the invention, the method for selecting the characteristic anchor point set A by the anchor point selection component comprises the following steps:
setting the number k of target anchor points to a crowd F i And F i+1 Dividing two groups of people into k subgroups by k-means clustering respectively, and then separating the distance from the gravity center of each subgroupThe nearest individual serves as an anchor point; center of gravity of subgroupThe components in each coordinate direction are:
X i is the position component in the X direction of any individual in the subgroup, Y i For the position component in the Y direction, and so on, W i Weight parameters for different individuals when calculating the center of gravity;
the anchor mapping component calculates an anchor set A i And A i+1 And then, taking the negative value of the distance between the individual world coordinate points in the two sets as a weight to establish a complete bipartite graph, and obtaining the optimal matching result of the total weight as the mapping relation between the anchor points.
According to the crowd formation planning system provided by the embodiment of the invention, the anchor point selection component acquires the Edge of the crowd and performs the crowd F i And F i+1 Edge of (2) i And Edge i+1 Respectively carrying out down-sampling to ensure Edge after sampling' i And Edge' i+1 The equipotential is satisfied:
Edge′ i ~Edge′ i+1
will Edge' i And Edge' i+1 As a population F i And F i+1 The anchor point set of (2);
edge 'since the crowd Edge is continuous' i And Edge' i+1 Can be regarded as two circular queues with equal length, the anchor point mapping component designates a certain agent to map: edge' i (j)→Edge′ i+1 (k) And the other agents complete the mapping according to the sequence in the queue.
According to the crowd formation planning system provided by the embodiment of the invention, the anchor point mapping component obtains the optimal matching result of the total weight value through a Kuhn-Munkres algorithm to be used as the mapping relation between anchor points.
The invention also provides a terminal, which comprises a processor, an input device, an output device and a memory, wherein the memory is used for storing program codes, and the processor is configured to run the program codes and execute the crowd formation planning method.
The invention also provides a computer-readable storage medium having stored thereon program instructions which, when executed by a processor, cause the processor to carry out the aforementioned method of crowd formation planning.
Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art will be able to make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations fall within the scope of the invention defined by the appended claims.
Claims (10)
1. A crowd formation planning method based on feature anchor points is characterized by comprising the following steps:
generating N frames of patterns corresponding to the crowd arrangement shape according to a time sequence;
secondly, determining the point location distribution of the crowd corresponding to each frame of pattern;
step three, determining a mapping scheme from the ith frame to the (i + 1) th frame pattern crowd point position, wherein the specific method comprises the following steps:
3.2, determining the mapping relation of anchor point sets in the two frames of patterns;
3.3, defining a loss function of the crowd formation transformation according to the anchor point:
for the population F i Each non-anchor individual agent of ∈ A' i Then the population F i Position p in i To the crowd F i+1 Middle position p i+1 Distance loss D of loss And the azimuth loss O loss Is defined as follows:
defining a loss function:
V loss =εD loss +λO loss +μD
wherein epsilon, lambda and mu are proportional parameters, D is the measurement space distance from agt to agt' in the traditional method,
3.4, obtaining F i All agent pairs F i+1 V of all positions loss Obtaining a macroscopic planning result with the lowest total loss;
and step four, obtaining the movement direction and speed of each person according to the mapping scheme.
2. The crowd formation planning method according to claim 1, wherein the method for selecting the feature anchor point set a in step 3.1 comprises:
setting the number k of target anchor points to the crowd F i And F i+1 Dividing two groups of people into k subgroups by k-means clustering respectively, and then separating the distance from the gravity center of each subgroupThe nearest individual serves as an anchor point; center of gravity of subgroupThe components in each coordinate direction are:
X i is the position component in the X direction of any individual in the subgroup, Y i For the position component in the Y direction, and so on, W i Calculating weight parameters of different individuals when calculating the gravity center;
the method for determining the anchor point set mapping relation in the two-frame pattern in the step 3.2 comprises the following steps:
anchor point set A is obtained through calculation i And A i+1 And then, taking the negative value of the distance between the individual world coordinate points in the two sets as a weight to establish a complete bipartite graph, and obtaining the optimal matching result of the total weight as the mapping relation between the anchor points.
3. The population formation planning method according to claim 1, wherein the method for selecting the characteristic anchor point set a and determining the mapping relationship of the anchor point set in step 3.1 and step 3.2 comprises:
acquiring the Edge of a crowd; for the crowd F i And F i+1 Edge of (2) i And Edge i+1 Respectively carrying out down-sampling to ensure Edge after sampling' i And Edge' i+1 The equipotential is satisfied:
Edge′ i ~Edge′ i+1
specifying one of the agent mappings: edge' i (j)→Edge′ i+1 (k) And the other agents complete mapping according to the sequence in the queue.
4. The crowd formation planning method based on claim 2, wherein the optimal matching result of the total weight is obtained through a Kuhn-Munkres algorithm and is used as a mapping relation between anchor points.
5. A crowd formation planning system based on feature anchor points is characterized by comprising:
the pattern generating module is used for generating N frames of patterns corresponding to the crowd arrangement shape according to the time sequence;
the point location setting module is used for determining the point location distribution of the crowd corresponding to each frame of pattern;
the point location mapping module is used for determining a mapping scheme from the ith frame to the (i + 1) th frame pattern crowd point location;
the motion conversion module is used for obtaining the motion direction and the motion speed of each person according to the mapping scheme;
wherein, the point location mapping module includes:
anchor selection means for selecting a set of characteristic anchors in the population FThe non-anchor set is A' = F-A;
the anchor point mapping component is used for determining the mapping relation of the anchor point set in the two frames of patterns;
a loss function component for defining a loss function of the population formation transform according to the anchor point:
for the population F i Each non-anchor individual agent of ∈ A' i Then the population F i Position p in i To the crowd F i+1 Middle position p i+1 Distance loss D of loss And the azimuth loss O loss Is defined as follows:
defining a loss function:
V loss =εD loss +λO loss +μD
wherein epsilon, lambda and mu are proportional parameters, D is the measurement space distance from agt to agt' in the traditional method,
crowd planning means for obtaining F i All agent pairs F in i+1 V of all positions loss And obtaining a macroscopic planning result with the lowest total loss.
6. The crowd formation planning system of claim 5, wherein the anchor selection component selects the feature anchor set A by:
setting the number k of target anchor points to the crowd F i And F i+1 Dividing two groups of people into k subgroups by k-means clustering respectively, and then separating the distance from the gravity center of each subgroupThe nearest individual serves as an anchor point; center of gravity of subgroupThe components in each coordinate direction are:
X i is the position component in the X direction of any individual in the subgroup, Y i For the position component in the Y direction, and so on, W i Calculating weight parameters of different individuals when calculating the gravity center;
the anchor point mapping component calculates to obtain an anchor point set A i And A i+1 And then, taking the negative value of the distance between the individual world coordinate points in the two sets as a weight to establish a complete bipartite graph, and obtaining the optimal matching result of the total weight as the mapping relation between anchor points.
7. The crowd formation planning system of claim 5, wherein the anchor point selection component obtains the Edge of the crowd, and for the crowd F i And F i+1 Edge of (2) i And Edge i+1 Respectively carrying out down-sampling to ensure Edge after sampling' i And Edge' i+1 The equipotential is satisfied:
Edge′ i ~Edge′ i+1
will Edge' i And Edge' i+1 As a population F i And F i+1 The set of anchor points of (a);
edge 'since the crowd Edge is continuous' i And Edge' i+1 Can be regarded as two circular queues with equal length, the anchor point mapping component designates a certain agent to map: edge' i (j)→Edge′ i+1 (k) And the other agents complete mapping according to the sequence in the queue.
8. The crowd formation planning system according to claim 6, wherein the anchor point mapping component obtains the optimal matching result of the total weight as the mapping relationship between anchor points through a Kuhn-Munkres algorithm.
9. A terminal comprising a processor, an input device, an output device and a memory, wherein the memory is configured to store program code, and the processor is configured to execute the program code to perform the human motion delta identification method according to any of claims 1-4.
10. A computer-readable storage medium, characterized in that the computer storage medium stores program instructions that, when executed by a processor, cause the processor to perform the human motion increment recognition method according to any one of claims 1 to 4.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211073984.7A CN115358621A (en) | 2022-09-02 | 2022-09-02 | Crowd formation planning method and system based on characteristic anchor points |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211073984.7A CN115358621A (en) | 2022-09-02 | 2022-09-02 | Crowd formation planning method and system based on characteristic anchor points |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115358621A true CN115358621A (en) | 2022-11-18 |
Family
ID=84007516
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211073984.7A Pending CN115358621A (en) | 2022-09-02 | 2022-09-02 | Crowd formation planning method and system based on characteristic anchor points |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115358621A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117372946A (en) * | 2023-09-19 | 2024-01-09 | 日照市规划设计研究院集团有限公司 | Tourist group tourist behavior identification method |
-
2022
- 2022-09-02 CN CN202211073984.7A patent/CN115358621A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117372946A (en) * | 2023-09-19 | 2024-01-09 | 日照市规划设计研究院集团有限公司 | Tourist group tourist behavior identification method |
CN117372946B (en) * | 2023-09-19 | 2024-04-16 | 日照市规划设计研究院集团有限公司 | Tourist group tourist behavior identification method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP3944200A1 (en) | Facial image generation method and apparatus, device and storage medium | |
CN108416840A (en) | A kind of dense method for reconstructing of three-dimensional scenic based on monocular camera | |
KR101555347B1 (en) | Apparatus and method for generating video-guided facial animation | |
CN109583509B (en) | Data generation method and device and electronic equipment | |
CN109190461B (en) | A kind of dynamic gesture identification method and system based on gesture key point | |
CN101877146B (en) | Method for extending three-dimensional face database | |
CN110246181A (en) | Attitude estimation model training method, Attitude estimation method and system based on anchor point | |
CN109389156B (en) | Training method and device of image positioning model and image positioning method | |
CN114385376B (en) | Client selection method for federal learning of lower edge side of heterogeneous data | |
CN108961385B (en) | SLAM composition method and device | |
CN111815768B (en) | Three-dimensional face reconstruction method and device | |
CN110443285A (en) | The determination method, apparatus and computer storage medium of similar track | |
CN112232134A (en) | Human body posture estimation method based on hourglass network and attention mechanism | |
CN115358621A (en) | Crowd formation planning method and system based on characteristic anchor points | |
CN105976395A (en) | Video target tracking method based on sparse representation | |
Yamasaki et al. | Motion segmentation and retrieval for 3D video based on modified shape distribution | |
CN108846845A (en) | SAR image segmentation method based on thumbnail and hierarchical fuzzy cluster | |
CN106022293B (en) | A kind of pedestrian's recognition methods again based on adaptive sharing niche evolution algorithm | |
CN117216591A (en) | Training method and device for three-dimensional model matching and multi-modal feature mapping model | |
CN115482557B (en) | Human body image generation method, system, equipment and storage medium | |
CN108614889B (en) | Moving object continuous k nearest neighbor query method and system based on Gaussian mixture model | |
CN111275610A (en) | Method and system for processing face aging image | |
CN111080517B (en) | Three-dimensional point cloud splicing method based on improved butterfly optimization algorithm | |
CN109711363A (en) | Vehicle positioning method, device, equipment and storage medium | |
CN116704097B (en) | Digitized human figure design method based on human body posture consistency and texture mapping |
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 |