CN116310243B - AR anti-shake method, system, equipment and storage medium based on artificial intelligence - Google Patents
AR anti-shake method, system, equipment and storage medium based on artificial intelligence Download PDFInfo
- Publication number
- CN116310243B CN116310243B CN202310586577.4A CN202310586577A CN116310243B CN 116310243 B CN116310243 B CN 116310243B CN 202310586577 A CN202310586577 A CN 202310586577A CN 116310243 B CN116310243 B CN 116310243B
- Authority
- CN
- China
- Prior art keywords
- model
- size
- artificial intelligence
- simple model
- shake
- 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
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 37
- 238000000034 method Methods 0.000 title claims abstract description 25
- 230000000007 visual effect Effects 0.000 claims abstract description 35
- 238000007906 compression Methods 0.000 claims abstract description 11
- 230000006835 compression Effects 0.000 claims abstract description 11
- 238000006243 chemical reaction Methods 0.000 claims abstract description 4
- 238000012217 deletion Methods 0.000 claims abstract description 3
- 230000037430 deletion Effects 0.000 claims abstract description 3
- 238000013527 convolutional neural network Methods 0.000 claims description 28
- 210000002569 neuron Anatomy 0.000 claims description 20
- 238000013507 mapping Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000000694 effects Effects 0.000 abstract description 6
- 238000005094 computer simulation Methods 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 4
- 230000003190 augmentative effect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 238000013144 data compression Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
- G06T19/006—Mixed reality
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
- G06T19/20—Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Graphics (AREA)
- General Engineering & Computer Science (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Architecture (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Processing Or Creating Images (AREA)
Abstract
The application relates to the technical field of computer models applying artificial intelligence, in particular to an AR anti-shake method, an AR anti-shake system, AR anti-shake equipment and an AR anti-shake storage medium based on artificial intelligence, wherein the AR anti-shake method comprises the following steps: constructing a three-dimensional model based on a pre-acquired physical image; model compression is carried out based on the three-dimensional model, and a simple model is established; according to the space positioning of the AR camera, the position matching of the simple model center point and the real object center point is carried out, and the specific position for generating the simple model is determined; calculating the size of the simple model based on artificial intelligence, and determining the size proportion of the generated simple model; based on the specific position and the size proportion of the generated simple model, model loading is carried out in an AR scene, and the deletion or the conversion of the simple model into a map is carried out according to the positioning point and the visual radius of the AR camera in the virtual space. The loading speed of the model in the scene is improved, so that the model is stably attached to a real object, shake is prevented, and the effects of low power consumption and low delay are effectively achieved.
Description
Technical Field
The application relates to the technical field of computer models applying artificial intelligence, in particular to an AR anti-shake method, an AR anti-shake system, AR anti-shake equipment and an AR anti-shake storage medium based on artificial intelligence.
Background
AR augmented reality (Augmented Reality, abbreviated as AR) is a technology for calculating the position and angle of a camera image in real time and adding a corresponding image, and is a technology for nesting a virtual world in the real world on a screen and performing interaction.
At present, in the AR field, jitter is always a problem under discussion, when a user wears an AR camera to experience a displayed three-dimensional model, the user usually drops frames due to the frame rate of the content of the model, is blocked and unstable during playing, and causes jitter due to the fact that when the scene is identified in real time, the quality of an acquired scene identification picture is not unique, and the scene is subject to jitter caused by the fact that the difference of light and angles is changed. In addition, some recognition graphs are complex in modeling, delay phenomenon can occur when the model is loaded, so that errors exist in model data, and jitter phenomenon is caused.
Disclosure of Invention
In order to solve the problems, the application provides an AR anti-shake method, an AR anti-shake system, AR anti-shake equipment and an AR anti-shake storage medium based on artificial intelligence.
In a first aspect, the present application provides an artificial intelligence based AR anti-shake method applied to an AR camera, where the method includes the following steps:
constructing a three-dimensional model based on a pre-acquired physical image;
model compression is carried out based on the three-dimensional model, and a simple model is established;
according to the space positioning of the AR camera, the position matching of the simple model center point and the real object center point is carried out, and the specific position for generating the simple model is determined;
calculating the size of the simple model based on artificial intelligence, and determining the size proportion of the generated simple model;
based on the specific position and the size proportion of the generated simple model, model loading is carried out in an AR scene, and the deletion or the conversion of the simple model into a map is carried out according to the positioning point and the visual radius of the AR camera in the virtual space.
In some implementations of the first aspect, the model compression is performed based on a three-dimensional model, in particular, reducing a back surface, a shadow surface, or a puncture surface of the three-dimensional model.
In some implementations of the first aspect, the deleting of Jian Mo specifically includes:
and positioning according to the position of the AR camera in the virtual space, confirming the positioning point, and deleting the blocked simple model according to the model displayed in the visual range of the positioning point.
Further, the Jian Mo alternative is to locate the position of the AR camera in the virtual space, preset a visual radius with the locating point as the origin,
if the simple form is positioned within the visual radius, displaying the simple form;
if the simple form is located outside the visual radius, converting the simple form into a map for display;
and if the simple form is positioned on the visual radius, displaying the simple form.
In some implementations of the first aspect, the determining the specific location of the generated simple form is by obtaining a location O of a center point of the real object, and determining the location O of the center point of the simple form 1 Will position O 1 Is converted into world coordinates, O and O 1 And performing position matching so as to determine the specific position of the simple mode generation.
As a specific implementation mode, the artificial intelligence is a convolutional neural network, and neurons in each layer of the convolutional neural network are connected with neurons in the lower layer by adopting a quadtree data structure.
Further, the determining the size proportion of the generated simple mode includes:
taking the pre-acquired physical size as the input data size, wherein the size is W 1 ×H 1 ×D 1 ,W 1 To input the width of the data body H 1 To input the height of the data volume, D 1 Depth for the input data volume;
and define the output data size as W 2 ×H 2 ×D 2 The output data size is calculated as:
，
wherein W is 2 To output the width of the data body H 2 To output the height of the data volume, D 2 Depth for the output data volume; f is the receptive field size of the neurons in the convolutional neural network, S is the step size of the neurons in the convolutional neural network, K is the number of filters of the convolutional neural network, and P is the number of zero padding.
A second aspect provides an artificial intelligence based AR anti-shake system comprising:
the visual processing module is used for constructing a three-dimensional model based on a pre-acquired physical image;
the simple model building module is used for carrying out model compression based on the three-dimensional model to build a simple model;
the position matching module is used for matching the positions of the center point of the simple model and the center point of the real object according to the space positioning of the AR camera, and determining the specific position for generating the simple model;
the size calculation module is used for calculating the size of the simple model based on artificial intelligence and determining the size proportion of the generated simple model;
and the loading processing module is used for loading the model in the AR scene based on the specific position and the size proportion of the generated simple model, and deleting or converting the simple model into a map according to the positioning point and the visual radius of the AR camera in the virtual space.
A third aspect provides an artificial intelligence based AR anti-shake apparatus comprising a processor and a memory, wherein the processor implements an artificial intelligence based AR anti-shake method as described above when executing program data stored in the memory.
A fourth aspect provides a computer readable medium storing control program data, wherein the control program data, when executed by a processor, implements an artificial intelligence based AR anti-shake method as described above.
The beneficial effects are that: according to the application, the model compression is carried out on the three-dimensional model, so that the number of the model surfaces is reduced, the loading speed of the model in an AR scene is increased, and the delay effect is reduced; the convolutional neural network is adopted to calculate the size of the model, so that the model and the real object have the same space size, and the simple model and the real object can be stably attached; meanwhile, the position of the AR camera is used as a locating point, part of models are removed through shielding when the models are loaded, the number of simple models in the space is reduced, the speed of the AR camera for loading and displaying the simple models in a visual range is improved, the stability of display is improved, the shaking of the models is prevented, and the mapping conversion of the models is carried out according to the visual radius, so that the AR camera can better construct a virtual space, and the effects of low power consumption and low delay are achieved.
Drawings
FIG. 1 is a flow chart of an AR anti-shake method based on artificial intelligence of the application;
fig. 2 is a schematic structural diagram of an AR anti-shake system based on artificial intelligence according to the present application.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings.
As shown in fig. 1, the application provides an AR anti-shake method based on artificial intelligence, which comprises the following specific steps:
s1, constructing a three-dimensional model based on a pre-acquired physical image;
the method comprises the steps of carrying out identification processing on a pre-acquired physical image, extracting a plurality of feature vectors from the image, carrying out point cloud sample preparation on the extracted feature vectors through a three-dimensional engine, and constructing a three-dimensional model of a physical individual.
S2, performing model compression based on a three-dimensional model, and establishing a simple model;
and (3) performing model compression on the three-dimensional model by using a multsec (reduced model) optimization method, reducing the back surface, shadow surface or puncture surface of the three-dimensional model, reducing the number of model surfaces, and establishing a simple model of the three-dimensional model.
By establishing the simple model, the loading speed of the model in the AR scene is improved and the delay effect is reduced on the basis of not affecting the viewing experience.
S3, according to the space positioning of the AR camera, the position matching of the simple model center point and the real object center point is carried out, and the specific position for generating the simple model is determined;
acquiring a position O of a center point of a real object according to the space positioning of the AR camera and the data information acquired by the AR camera; at the same time, the position O of the simple mode central point is determined 1 And position O 1 Is converted into world coordinates, O and O 1 And performing position matching so as to determine the specific position of the simple mode generation.
S4, calculating the size of the simple model based on artificial intelligence, and determining the size proportion of the generated simple model;
taking the pre-acquired physical size as the input data size, wherein the size is W 1 ×H 1 ×D 1 ,W 1 To input the width of the data body H 1 To input the height of the data volume, D 1 Depth for the input data volume; and define the output data size as W 2 ×H 2 ×D 2 ，W 2 To output the width of the data body H 2 To output the height of the data volume, D 2 Depth for the output data volume;
as a specific implementation mode, the artificial intelligence adopts a convolutional neural network, and neurons in each layer of the convolutional neural network are connected with neurons in the lower layer through a quadtree data structure to perform size proportion calculation of simple model and real object. The convolutional neural network is provided with an input layer, a hidden layer and an output layer, the neurons in each layer of which are arranged in 3 dimensions, i.e. width, height and depth, wherein the depth of the convolutional neural network refers to the third dimension of the active data volume, not the depth of the entire network. Meanwhile, the layers of the convolutional neural network are connected by using a quadtree data structure, a quadtree (Q-Tree) is a Tree-shaped data structure, each node of the quadtree data structure can have at most four sub-nodes, a part of two-dimensional space is generally subdivided into four quadrants or areas, relevant information in the areas is stored in the quadtree nodes, and the areas can be square, rectangular or random.
When the convolutional neural network performs machine learning, the size of an input data body is input through an input layer, the size of an output data body is output from an output layer through processing of a hidden layer, meanwhile, if the actual output size of the output layer is larger than an expected value, counter-propagation is performed, errors are distributed to all neurons in the propagation process, and the weight of each neuron is modulated through error feedback of all neurons.
Further, the convolutional neural network is used for identification and analysis, the calculation of the size of the model output data volume is performed,
the size of the output data body is as follows:
，
wherein F is the receptive field size of the neurons in the convolutional neural network, S is the step length of the neurons in the convolutional neural network, K is the number of filters of the convolutional neural network, and P is the number of zero padding;
when the step length s=1, the zero padding value p= (F-1)/2, so that the input data body and the output data body can have the same space size, and the simple model can be stably attached to the real object.
S5, based on the specific position and the size proportion of the generated simple model, loading the model in the AR scene, and deleting or converting the simple model into a map according to the positioning point and the visual radius of the AR camera in the virtual space;
and (3) carrying out model loading by determining the specific position generated by the simple model and the calculated size of the output data body, and displaying in an AR scene.
Further, according to the positioning point of the AR in the virtual space, the simple mode displayed in the visual field range of the positioning point is determined, the blocked and rejected by Occlusion Culling (blocking and rejecting) technology is carried out, and the blocked simple mode is deleted in the AR scene, so that the number of the simple modes in the space is reduced, the loading and displaying speed of the AR camera to the simple modes in the visual field is improved, the displaying stability is improved, and the model shaking is prevented.
As a specific implementation mode, positioning is carried out according to the position of the AR camera in the virtual space, a positioning point is confirmed, and the blocked simple model is deleted according to the simple model displayed in the visible range of the positioning point; the boundary points of the simple models send detection rays to the positioning points, and if one or more boundary point detection rays of a certain model can directly irradiate the positioning points and are not overlapped or contacted with other simple models, the simple models are judged to be not blocked; on the contrary, judging that the simple module is blocked;
meanwhile, if the visual area of the defined simple model is smaller than an area display threshold, the defined simple model is defined as an occluded simple model, the area display threshold can be set in a self-defined mode, for example, the area display threshold can be set to be 5% of the area of the self-model, and if the area of the simple model which can be seen at a locating point is smaller than 5% of the volume of the self-model, the simple model is deleted in a scene.
Further, according to the positioning of the AR camera in the virtual space, a visual radius is determined, the positioning point of the position is used as an origin, a simple model outside the visual radius is converted into a mapping, and the mapping can be matched with the size of a real object by calculating the mapping size through the convolutional neural network in the step S4. Therefore, the AR camera can better construct a virtual space, and the effects of low power consumption and low delay are achieved.
Specifically, according to the position of the AR camera in the virtual space, a visual radius is preset by taking the positioning point as the origin,
if the simple form is positioned within the visual radius, displaying the simple form;
if the simple form is located outside the visual radius, converting the simple form into a map for display;
and if the simple form is positioned on the visual radius, displaying the simple form.
In addition, the present application provides an artificial intelligence based AR anti-shake system, the system comprising:
the visual processing module is used for constructing a three-dimensional model based on a pre-acquired physical image;
the simple model building module is used for carrying out model compression based on the three-dimensional model to build a simple model; and (3) carrying out data compression on the three-dimensional model by using a multsec (reduced mode) optimization method, reducing the back surface, shadow surface or puncture surface of the three-dimensional model, reducing the number of the modes, and establishing a simple mode of the three-dimensional model.
The position matching module performs position matching of a simple model center point and a real object center point according to the space positioning of the AR camera, and determines the specific position for generating the simple model; acquiring a position O of a center point of a real object according to the space positioning of the AR camera and the data information acquired by the AR camera; at the same time, the position O of the simple mode central point is determined 1 And position O 1 Is converted into world coordinates, O and O 1 And performing position matching so as to determine the specific position of the simple mode generation.
The size calculation module calculates the size of the simple model based on artificial intelligence, and determines the size proportion of the generated simple model;
the loading processing module is used for loading the model in the AR scene based on the specific position and the size proportion of the generated simple model, and deleting or converting the simple model into a map according to the positioning point and the visual radius of the AR camera in the virtual space;
the loading processing module determines a simple model displayed in the visual field range of the positioning point according to the positioning point of the AR in the virtual space, performs shielding and eliminating through a Occlusion Culling (shielding and eliminating) technology, and deletes the shielded simple model in an AR scene, so that the number of the simple models in the space is reduced, the loading display speed of the AR camera on the simple model in the visual field is improved, the display stability is improved, and model shaking is prevented;
the loading processing module is used for positioning according to the position of the AR camera in the virtual space, determining a visual radius, converting a simple model outside the visual radius into a mapping by taking the positioning point of the position as an origin, and matching the mapping with the size of a real object by calculating the mapping size through the convolutional neural network in the step S4. Therefore, the AR camera can better construct a virtual space, and the effects of low power consumption and low delay are achieved.
Furthermore, the application provides an AR anti-shake device based on artificial intelligence, which comprises a processor and a memory, wherein the processor realizes the AR anti-shake method based on artificial intelligence when executing program data stored in the memory.
Finally, the present application provides a computer readable medium storing control program data, wherein the control program data, when executed by a processor, implements the artificial intelligence based AR anti-shake method.
Claims (8)
1. An AR anti-shake method based on artificial intelligence is applied to an AR camera and is characterized by comprising the following steps:
constructing a three-dimensional model based on a pre-acquired physical image;
model compression is carried out based on the three-dimensional model, and a simple model is established;
according to the space positioning of the AR camera, the position matching of the simple model center point and the real object center point is carried out, and the specific position for generating the simple model is determined;
calculating the size of the simple model based on artificial intelligence, and determining the size proportion of the generated simple model; the artificial intelligence is a convolutional neural network, the neurons of each layer of the convolutional neural network are connected with the neurons in the lower layer by adopting a quadtree data structure,
taking the pre-acquired physical size as the input data size, wherein the size is W 1 ×H 1 ×D 1 ,W 1 To input the width of the data body H 1 To input the height of the data volume, D 1 For depth of input data volume, and defining output data size as W 2 ×H 2 ×D 2 The output data size is calculated as:
D 2 ＝K
wherein W is 2 To output the width of the data body H 2 To output the height of the data volume, D 2 Depth for the output data volume; f is the receptive field size of the neurons in the convolutional neural network, S is the step size of the neurons in the convolutional neural network, K is the number of filters of the convolutional neural network, and P is the number of zero padding;
based on the specific position and the size proportion of the generated simple model, model loading is carried out in an AR scene, and the deletion or the conversion of the simple model into a map is carried out according to the positioning point and the visual radius of the AR camera in the virtual space.
2. The AR anti-shake method according to claim 1, wherein the model compression is performed based on a three-dimensional model, in particular, reducing the back surface, shadow surface or puncture surface of the three-dimensional model.
3. The AR anti-shake method based on artificial intelligence according to claim 1, wherein the deleting of Jian Mo specifically comprises:
and positioning according to the position of the AR camera in the virtual space, confirming the positioning point, and deleting the blocked simple model according to the model displayed in the visual range of the positioning point.
4. The AR anti-shake method based on artificial intelligence according to claim 1, wherein the specific operation of replacing Jian Mo with mapping is as follows:
positioning according to the position of the AR camera in the virtual space, presetting a visual radius by taking the positioning point as an origin,
if the simple form is positioned within the visual radius, displaying the simple form;
if the simple form is located outside the visual radius, converting the simple form into a map for display;
and if the simple form is positioned on the visual radius, displaying the simple form.
5. The AR anti-shake method according to claim 1, wherein said determining a specific location of a generated simple is by obtaining a location O of a center point of a real object and determining a location O of a center point of a simple 1 Will position O 1 Is converted into world coordinates, O and O 1 And performing position matching so as to determine the specific position of the simple mode generation.
6. An artificial intelligence based AR anti-shake system, comprising:
the visual processing module is used for constructing a three-dimensional model based on a pre-acquired physical image;
the simple model building module is used for carrying out model compression based on the three-dimensional model to build a simple model;
the position matching module is used for matching the positions of the center point of the simple model and the center point of the real object according to the space positioning of the AR camera, and determining the specific position for generating the simple model;
the size calculation module is used for calculating the size of the simple model based on artificial intelligence and determining the size proportion of the generated simple model; the artificial intelligence is a convolutional neural network, the neurons of each layer of the convolutional neural network are connected with the neurons in the lower layer by adopting a quadtree data structure,
taking the pre-acquired physical size as the input data size, wherein the size is W 1 ×H 1 ×D 1 ,W 1 To input the width of the data body H 1 For inputting the height of the data volumeDegree, D 1 For depth of input data volume, and defining output data size as W 2 ×H 2 ×D 2 The output data size is calculated as:
D 2 ＝K
wherein W is 2 To output the width of the data body H 2 To output the height of the data volume, D 2 Depth for the output data volume; f is the receptive field size of the neurons in the convolutional neural network, S is the step size of the neurons in the convolutional neural network, K is the number of filters of the convolutional neural network, and P is the number of zero padding;
and the loading processing module is used for loading the model in the AR scene based on the specific position and the size proportion of the generated simple model, and deleting or converting the simple model into a map according to the positioning point and the visual radius of the AR camera in the virtual space.
7. An artificial intelligence based AR anti-shake apparatus comprising a processor and a memory, wherein the processor implements the artificial intelligence based AR anti-shake method of any of claims 1-5 when executing program data stored in the memory.
8. A computer readable medium for storing control program data, wherein the control program data, when executed by a processor, implements the artificial intelligence based AR anti-shake method according to any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310586577.4A CN116310243B (en) | 2023-05-24 | 2023-05-24 | AR anti-shake method, system, equipment and storage medium based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310586577.4A CN116310243B (en) | 2023-05-24 | 2023-05-24 | AR anti-shake method, system, equipment and storage medium based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116310243A CN116310243A (en) | 2023-06-23 |
| CN116310243B true CN116310243B (en) | 2023-08-15 |
Family
ID=86789188
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310586577.4A Active CN116310243B (en) | 2023-05-24 | 2023-05-24 | AR anti-shake method, system, equipment and storage medium based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116310243B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107451551A (en) * | 2017-07-24 | 2017-12-08 | 武汉秀宝软件有限公司 | A kind of optimization method and system for preventing float |
| CN110097645A (en) * | 2019-05-16 | 2019-08-06 | 中国人民解放军95927部队 | A kind of flight simulation emulated interface removal dither method and flight simulator |
| WO2020092477A1 (en) * | 2018-11-02 | 2020-05-07 | Facebook Technologies, Llc | Display engine for post-rendering processing |
| WO2021022962A1 (en) * | 2019-08-08 | 2021-02-11 | 华为技术有限公司 | Method and device for model inference based on graphics rendering pipelines, and storage medium |
| WO2022100685A1 (en) * | 2020-11-13 | 2022-05-19 | 华为技术有限公司 | Drawing command processing method and related device therefor |
| WO2023000745A1 (en) * | 2021-07-23 | 2023-01-26 | Oppo广东移动通信有限公司 | Display control method and related device |
| WO2023005355A1 (en) * | 2021-07-30 | 2023-02-02 | 荣耀终端有限公司 | Image anti-shake method and electronic device |
| CN116055856A (en) * | 2022-05-30 | 2023-05-02 | 荣耀终端有限公司 | Camera interface display method, electronic device, and computer-readable storage medium |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8860760B2 (en) * | 2010-09-25 | 2014-10-14 | Teledyne Scientific & Imaging, Llc | Augmented reality (AR) system and method for tracking parts and visually cueing a user to identify and locate parts in a scene |
| CN106663411A (en) * | 2014-11-16 | 2017-05-10 | 易欧耐特感知公司 | Systems and methods for augmented reality preparation, processing, and application |
| US10043319B2 (en) * | 2014-11-16 | 2018-08-07 | Eonite Perception Inc. | Optimizing head mounted displays for augmented reality |
| TWI526992B (en) * | 2015-01-21 | 2016-03-21 | 國立清華大學 | Method for optimizing occlusion in augmented reality based on depth camera |
-
2023
- 2023-05-24 CN CN202310586577.4A patent/CN116310243B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107451551A (en) * | 2017-07-24 | 2017-12-08 | 武汉秀宝软件有限公司 | A kind of optimization method and system for preventing float |
| WO2020092477A1 (en) * | 2018-11-02 | 2020-05-07 | Facebook Technologies, Llc | Display engine for post-rendering processing |
| CN110097645A (en) * | 2019-05-16 | 2019-08-06 | 中国人民解放军95927部队 | A kind of flight simulation emulated interface removal dither method and flight simulator |
| WO2021022962A1 (en) * | 2019-08-08 | 2021-02-11 | 华为技术有限公司 | Method and device for model inference based on graphics rendering pipelines, and storage medium |
| WO2022100685A1 (en) * | 2020-11-13 | 2022-05-19 | 华为技术有限公司 | Drawing command processing method and related device therefor |
| WO2023000745A1 (en) * | 2021-07-23 | 2023-01-26 | Oppo广东移动通信有限公司 | Display control method and related device |
| WO2023005355A1 (en) * | 2021-07-30 | 2023-02-02 | 荣耀终端有限公司 | Image anti-shake method and electronic device |
| CN116055856A (en) * | 2022-05-30 | 2023-05-02 | 荣耀终端有限公司 | Camera interface display method, electronic device, and computer-readable storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 葛良水 ; 胡少华 ; 商莹 ; .基于ARToolKit的二维码多标识增强现实系统.机械设计与制造工程.2018,(06),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116310243A (en) | 2023-06-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114004941B (en) | Indoor scene three-dimensional reconstruction system and method based on nerve radiation field | |
| CN108335353B (en) | Three-dimensional reconstruction method, device and system of dynamic scene, server and medium | |
| CN112150575B (en) | Scene data acquisition method, model training method and device and computer equipment | |
| CN111369681B (en) | Three-dimensional model reconstruction method, device, equipment and storage medium | |
| CN108898630A (en) | A kind of three-dimensional rebuilding method, device, equipment and storage medium | |
| CN107223269A (en) | Three-dimensional scene positioning method and device | |
| US20020095276A1 (en) | Intelligent modeling, transformation and manipulation system | |
| CN111415420B (en) | Spatial information determining method and device and electronic equipment | |
| CN116977522A (en) | Rendering method and device of three-dimensional model, computer equipment and storage medium | |
| CN112749611B (en) | Face point cloud model generation method and device, storage medium and electronic equipment | |
| US11989900B2 (en) | Object recognition neural network for amodal center prediction | |
| CN110717967A (en) | Large-scene-model-oriented web-side dynamic rendering LOD processing method | |
| CN108550181B (en) | Method, system and equipment for online tracking and dense reconstruction on mobile equipment | |
| CN112102342A (en) | Plane contour recognition method and device, computer equipment and storage medium | |
| CN116310243B (en) | AR anti-shake method, system, equipment and storage medium based on artificial intelligence | |
| CN116109803B (en) | Information construction method, device, equipment and storage medium | |
| CN111870953A (en) | Height map generation method, device, equipment and storage medium | |
| CN115953541A (en) | Quad-tree LOD terrain generation method, device, equipment and storage medium | |
| JPH10293862A (en) | Three-dimensional object data processing method | |
| CN115393538A (en) | Visual SLAM method and system for indoor dynamic scene based on deep learning | |
| JPH0636013A (en) | Method and device for generating topographic data | |
| JPH10198823A (en) | Video generating device | |
| CN112802175A (en) | Large-scale scene occlusion rejection method, device, equipment and storage medium | |
| CN117611781B (en) | Flattening method and device for live-action three-dimensional model | |
| CN116258822B (en) | Three-dimensional engine boundary defining method, device and storage medium based on meta universe |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |