CN111080776A - Processing method and system for human body action three-dimensional data acquisition and reproduction - Google Patents
Processing method and system for human body action three-dimensional data acquisition and reproduction Download PDFInfo
- Publication number
- CN111080776A CN111080776A CN201911319622.XA CN201911319622A CN111080776A CN 111080776 A CN111080776 A CN 111080776A CN 201911319622 A CN201911319622 A CN 201911319622A CN 111080776 A CN111080776 A CN 111080776A
- Authority
- CN
- China
- Prior art keywords
- human body
- dimensional
- model
- motion
- data acquisition
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- 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
- G06T19/006—Mixed reality
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention relates to a processing method and a system for acquiring and reproducing three-dimensional data of human body actions, wherein the method comprises the following steps: acquiring a human body action image comprising a plurality of different angles; generating mapping information; carrying out three-dimensional human body model reconstruction on the image data set; extracting bones of the reconstructed three-dimensional human body model by using a bone extraction algorithm to obtain a three-dimensional human body model bound with the bones; acquiring a human body motion video stream, extracting human body actions through a deep learning algorithm, and fusing the extracted human body actions into a three-dimensional human body model bound with bones to generate a human body model with actions; the three-dimensional human body action is reproduced. The system comprises an image data acquisition unit, a human body model three-dimensional reconstruction unit, a human body action model generation unit, a human body action three-dimensional model reproduction unit and a data storage unit. The invention improves the accuracy of the model for calculating the human body actions and provides necessary support for the acquisition and the reproduction of various human body actions.
Description
Technical Field
The invention relates to the technical field of computer vision processing, in particular to a processing method and a system for acquiring and reproducing three-dimensional data of human body actions.
Background
Dynamic models of virtual characters are an important component of modern industry development, such as smart factories, movie and game visual effect applications, etc., with immeasurable value for reality, immersive virtual and augmented reality, telepresence, and three-dimensional, and free viewpoint video. The ideal characteristics of such virtual characters are advanced, unified, dense shapes, motion and variation models, and human appearance, regardless of the physical constitution or fashion of the garment. Authoring such models with high fidelity typically requires months of work by a professional. To simplify this process, we have investigated methods for the acquisition and reconstruction of three-dimensional data of body motion without markers, reconstructing a portion of such a three-dimensional model from camera recordings of real body motion.
Three-dimensional reconstruction refers to the establishment of a mathematical model suitable for computer representation and processing of a three-dimensional object, is the basis for processing, operating and analyzing the properties of the three-dimensional object in a computer environment, and is also a key technology for establishing virtual reality expressing an objective world in a computer. In computer vision, three-dimensional reconstruction refers to the process of reconstructing three-dimensional information from single-view or multi-view images. Since the information of a single video is incomplete, the three-dimensional reconstruction needs to utilize empirical knowledge. The method is that the camera is calibrated, namely the relation between the image coordinate system of the camera and the world coordinate system is calculated, and then three-dimensional information is reconstructed by utilizing the information in a plurality of two-dimensional images.
Currently, three-dimensional reconstruction apparatuses or methods commonly used in the market include three types. The first method is based on laser scanning, which is mainly based on the principle of laser ranging, and the construction of such devices mainly includes a high-speed precise laser range finder, a digital camera, and an auxiliary device for guiding the laser movement. The method adopting laser scanning has the advantages of high precision and high speed, but the equipment is often expensive and the volume and the like are often large; the second method is a method based on a single camera and auxiliary information, and such a method needs to print auxiliary information such as active light spots on an object and calculate three-dimensional position information of the object by analyzing changes such as deformation of texture light spots, for example, some devices project structured light spot information to the object by using an infrared projector. This type of method has the disadvantage of requiring additional auxiliary equipment and can be difficult in scenes where objects are moving rapidly. In addition, some three-dimensional reconstruction methods are available, the object contour is not completely segmented, the object contour and the object appearance color information are not accurately obtained, the calculation amount is large, and the steps are complex. The third method is a method based on stereoscopic vision, that is, an object is shot from different angles by more than two cameras, image data obtained is analyzed, and spatial positions of three-dimensional points of the object are calculated from the image data, so that three-dimensional reconstruction data is obtained.
Disclosure of Invention
Aiming at the defects and defects of the prior art, the invention provides a processing method and a system for acquiring and reproducing three-dimensional data of human body actions.
In order to achieve the purpose, the technical scheme of the invention is as follows:
a processing method for human body action three-dimensional data acquisition and reproduction is characterized by comprising the following steps: the method comprises the following steps:
s1, acquiring human body images including a plurality of different angles;
s2, carrying out space matching processing on human body image information of a plurality of different angles, then fusing the processed image information into a point cloud model, and then carrying out feature matching and cutting on the image to generate mapping information;
s3, triangularization processing and BA optimization are carried out on the generated point cloud model data with the mapping information to generate an image data set, and three-dimensional human body model reconstruction is carried out on the image data set;
s4, extracting bones of the three-dimensional human body model reconstructed in the step S3 by using a bone extraction algorithm to obtain a three-dimensional human body model bound with the bones;
s5, obtaining a human motion video stream, extracting human motion through a deep learning algorithm, adding the extracted human motion into the three-dimensional human model bound with bones in the step S4, and generating a human model with motion;
and S6, reproducing the three-dimensional human body motion.
Further, in S1, distortion correction processing needs to be performed on the acquired human body image.
Further, the human body image in S1 is captured by a plurality of high-definition cameras.
Furthermore, in S1, when the plurality of high definition cameras capture the human body image, reference points of the capture terminal need to be calibrated, and the spatial orientation and the internal parameters of the capture terminal are determined.
Further, the bone extraction algorithm in S4 adopts a K3M sequential iteration algorithm.
Further, in S6, the human body model with motion generated in S5 is input to the VR reconstruction system, and the three-dimensional human body motion is reconstructed by the VR reconstruction system.
Further, the human motion image acquired in S1 and the human motion video stream acquired in S5 may be simultaneously obtained by different or the same photographing apparatuses.
A processing system for human body motion three-dimensional data acquisition and reproduction is characterized by comprising:
the image data acquisition unit is used for acquiring human body images at a plurality of different angles;
the human body model three-dimensional reconstruction unit is used for performing three-dimensional human body model reconstruction on the human body image acquired by the image data acquisition unit and extracting bones of the reconstructed three-dimensional human body model so as to obtain a three-dimensional human body model bound with the bones;
the human body action model generating unit is used for acquiring a human body motion video stream, extracting human body actions through a deep learning algorithm, adding the extracted human body actions into a three-dimensional human body model bound with bones, and generating a human body action model with actions;
the human body action three-dimensional model reproduction unit is used for reproducing the human body action after the human body action model is finished;
and the data storage unit is used for storing and interacting the data acquired by the image data acquisition unit, the data between the human body model three-dimensional reconstruction unit and the human body action model generation unit.
Further, the image data acquisition unit in the system comprises a shooting terminal consisting of a plurality of cameras capable of shooting simultaneously.
Further, the human body action three-dimensional model reproduction unit in the system comprises a VR reproduction display system.
The technical scheme has the following beneficial effects: compared with the traditional human body three-dimensional modeling method, the human body three-dimensional modeling is intensively and quickly completed through multi-angle character image synchronous acquisition, the human body posture estimation is carried out through deep learning based on video information by shooting a human body motion video, the human body motion is extracted, the three-dimensional reconstruction is carried out by combining the three-dimensional human body model bound with bones, the human body model with motion is generated and is virtually displayed, the whole process of the human body motion is reproduced, the accuracy of the model calculation human body motion is improved through the deep learning method, and necessary support is provided for the acquisition and reproduction of various human body motions.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Detailed Description
To facilitate an understanding of the invention, the invention will now be described more fully with reference to the accompanying drawings. Preferred embodiments of the present invention are shown in the drawings. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The use of the terms herein are exemplary only, and are not limiting. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.
As shown in fig. 1, a processing method for acquiring and reproducing three-dimensional data of human body actions is characterized in that: the method comprises the following steps:
and S1, acquiring human body images including a plurality of different angles, wherein the human body images of the plurality of different angles are acquired by shooting through a plurality of high-definition cameras, and acquiring the spatial orientation and internal parameters of the plurality of high-definition cameras at the same time of shooting, wherein the spatial orientation is the orientation of the camera in the world coordinate system, and the orientation information of the camera in the world coordinate system is acquired for distortion correction. The internal parameters include the focal length of the camera, the camera principal point coordinates, and imaging distortion parameters. A plurality of high definition digtal cameras of this embodiment are set up in a three-dimensional shooting space, this space belongs to a hollow cubical space, position equipartition is moved about freely and quickly all around in this space has a plurality of high definition digtal cameras, for example 20 shooting poles are installed to a cylinder cubical space fixed interval position, evenly arrange 5 cameras on every pole, install 100 altogether, and, moreover, still can set up control lighting apparatus in the cubical space, it has sufficient light source to guarantee to shoot the space, be used for promoting the shooting effect. In addition, a plurality of cameras can be connected to the same control switch, and a key of the plurality of cameras is used for simultaneously shooting and acquiring images, and it needs to be added that distortion correction processing needs to be performed on the acquired human body images after the human body images are acquired, and the distortion process refers to the following steps:
the camera needs to be calibrated before the computer vision detection and identification, the camera is calibrated through a camera calibration tool in Matlab to simply calibrate radial distortion and tangential distortion, and three radial distortion parameters k1, k2 and k3 and two tangential distortion parameters p1 and p2 of the camera are obtained. These five parameters can be used in subsequent opencv image correction.
The radial distortion can be corrected by the following taylor series expansion:
xcorrected1＝x(1+k1r2+k2r4+k3r6)
ycorrected1＝y(1+k1r2+k2r4+k3r6)
here, x and y are distorted position coordinates in the image, and the true coordinates are obtained by correction. r is the distance of the point from the imaging center.
The tangential distortion can be corrected by the following equation:
xcorrected2＝x+[2p1y+p2(r2+2x2)]
ycorrected2＝y+[2p1x+p2(r2+2y2)]
here, x and y are distorted position coordinates in the image, and the true coordinates are obtained by correction. r is the distance of the point from the imaging center.
In the embodiment, an opencv distortion model is adopted for distortion correction.
Wherein the radial distortion model is: derived from Taylor's equation, in opencv K is 1, r2＝x2+y2(x, y) are real coordinates (distortion occurs), (x ', y') are ideal coordinates.
δxr＝x(k1r2+k2r4+k3r6+K)
δyr＝y(k1r2+k2r4+k3r6+K)
The tangential distortion model is:
δxd＝2p1xy+p2(r2+2x2)+K
δyd＝2p1(r2+2y2)+2p2xy+K
ideal coordinates (x ', y') and real coordinates (x, y):
x’＝x+δxr+δxd
y’＝y+δyr+δyd
the following can be obtained:
the distortion obtained by the formula has five parameters of k1, k2, k3, p1 and p2, the tangential distortion is small and negligible for a camera with better quality, the radial distortion coefficient k3 can also be ignored, and only two parameters of k1 and k2 are calculated.
S2, carrying out space matching processing on human body image information of a plurality of different angles, then fusing the processed image information into a point cloud model, and then carrying out feature matching and cutting on the image to generate mapping information;
s3, triangularization processing and BA optimization are carried out on the generated point cloud model data with the mapping information to generate an image data set, and three-dimensional human body model reconstruction is carried out on the image data set;
and S4, extracting bones of the three-dimensional human body model reconstructed in the step S3 by utilizing a bone extraction algorithm to obtain the three-dimensional human body model bound with the bones, wherein the bone extraction algorithm adopts a K3M sequential iteration algorithm, specifically, a target peripheral contour is firstly extracted, and then a target image boundary is corroded by utilizing the contour (which is an iteration process) until the boundary is corroded to be incapable of being corroded any more. The specific algorithm is divided into two blocks:
and (3) continuously corroding the first block to extract a pseudo skeleton (part of the area has two layers of pixel widths, but is close to the real skeleton). And a second block, extracting a real skeleton from the pseudo skeleton.
Wherein the first block: and extracting the pseudo skeleton, wherein each iteration is carried out for 6 steps, and the iteration is continuously carried out until no new pixel is corroded in the target boundary in the process of a certain iteration, and at the moment, the residual target image pixels are very close to the real skeleton (called as the pseudo skeleton).
First step in first block: the latest target contour (initially the original target contour) is extracted and these contour points are recorded.
And secondly, sequentially detecting the 8-pixel neighborhoods of the contour points, judging whether only 3 connected pixels exist, if so, deleting the points from the contour points, and deleting (corroding) corresponding points in the target image.
And thirdly, sequentially detecting whether the 8-pixel neighborhood of the rest contour points in the second step only contains 3 or 4 connected pixels, if so, deleting the points from the contour points, and deleting (corroding) corresponding points in the target image.
And fourthly, sequentially detecting whether the 8-pixel neighborhood of the rest contour points in the third step only contains 3 or 4 or 5 connected pixels, if so, deleting the points from the contour points, and deleting (corroding) corresponding points in the target image.
And fifthly, sequentially detecting whether the 8-pixel neighborhood of the residual contour points in the fourth step only contains 3 or 4 or 5 or 6 connected pixels, if so, deleting the points from the contour points, and deleting (corroding) corresponding points in the target image.
And sixthly, sequentially detecting whether the 8-pixel neighborhood of the residual contour points in the fifth step only contains 3 or 4 or 5 or 6 or 7 connected pixels, if so, deleting the points from the contour points, and deleting (corroding) corresponding points in the target image. This is the last step of an iterative process, and if there are still pixels eroded in this step, it means that there is "flesh" in addition to the real skeleton, and the boundary still needs to be eroded. This step is the termination condition for the iteration of the algorithm.
And a second block, extracting a real skeleton from the pseudo skeleton, wherein a specific algorithm is that a partial region of the pseudo skeleton obtained according to the first block has two pixel widths, and the target skeleton has a single-layer pixel width. Therefore, the final skeleton is extracted by the following steps:
and sequentially detecting whether the 8-pixel neighborhood of the pseudo skeleton in the target image only contains 2, 3, 4, 5, 6 or 7 connected pixels, and if so, deleting (corroding) the point from the pseudo skeleton to obtain the final skeleton.
S5, obtaining a human motion video stream, extracting human motion through a deep learning algorithm, adding the extracted human motion into the three-dimensional human model bound with bones in the step S4, and generating a human model with motion; the Deep learning (Deep learning) of the embodiment adopts a new network structure, namely, a standard Hourglass network, to recognize the posture of the human body, and the network structure can capture and integrate information of all scales of the image. The information under each scale is captured through the Hourglass module, for example, local features are needed when parts such as faces and hands are captured, and overall information needs to be captured when the human body posture is predicted. In order to capture the features of a picture at multiple scales, it is common practice to process the information at different scales separately using multiple pipeline, and then to combine the features at the rear part of the network, specifically to save the spatial information at each scale by using a single pipeline with skip layers.
And S6, inputting the human body model with the motion generated in the S5 into a VR reproduction system, and reproducing the three-dimensional human body motion through the VR display system.
It should be noted that the human motion image acquired in S1 and the human motion video stream acquired in S5 may be acquired by different or the same photographing apparatuses at the same time.
A processing system for human body motion three-dimensional data acquisition and reproduction comprises:
the image data acquisition unit is used for acquiring human body images at a plurality of different angles; the image data acquisition unit is realized by a plurality of high-definition cameras in a hollow three-dimensional space.
The human body model three-dimensional reconstruction unit is used for performing three-dimensional human body model reconstruction on the human body image acquired by the image data acquisition unit and extracting bones of the reconstructed three-dimensional human body model so as to obtain a three-dimensional human body model bound with the bones;
the human body action model generating unit is used for acquiring a human body motion video stream, extracting human body actions through a deep learning algorithm, adding the extracted human body actions into a three-dimensional human body model bound with bones, and generating a human body action model with actions;
the human body model three-dimensional reconstruction unit and the human body motion model generation unit can be realized centrally by a processor integrated with corresponding processing software, the processor is not limited to a central processing unit, other general processors, a digital signal processor, an application-specific integrated circuit, an off-the-shelf programmable gate array or other programmable logic devices, a discrete gate or transistor logic device, a discrete hardware component, etc., the general processor can be a microprocessor or the processor can also be any conventional processor, etc., the processor is a control center of the shooting device and the output device, and various interfaces and lines are used for connecting each terminal shooting device and the output device, etc.
The human body action three-dimensional model reproduction unit is used for reproducing the human body action after the human body action model is finished; the human body action three-dimensional model reproduction unit adopts a VR display system and is connected with the processor through a corresponding interface.
The data storage unit adopts an external memory or an internal memory which is connected with the processor and is used for storing and interacting the data acquired by the image data acquisition unit, the data between the human body model three-dimensional reconstruction unit and the human body action model generation unit, wherein the memory can be used for storing computer programs and/or modules and/or data, and the processor realizes various functions by running or executing the computer programs and/or modules stored in the memory and calling the data stored in the memory. In addition, the memory may be a high speed random access memory, but may also be a non-volatile, tangible memory such as a hard disk, a memory, a flash memory card, or other volatile solid state memory device.
The above embodiments are merely representative of the centralized embodiments of the present invention, and the description thereof is specific and detailed, but it should not be understood as the limitation of the scope of the present invention, and it should be noted that those skilled in the art can make various changes and modifications without departing from the spirit of the present invention, and these changes and modifications all fall into the protection scope of the present invention. Therefore, the protection scope of the present patent should be subject to the appended claims.
Claims (10)
1. A processing method for human body action three-dimensional data acquisition and reproduction is characterized by comprising the following steps: the method comprises the following steps:
s1, acquiring human body images including a plurality of different angles;
s2, carrying out space matching processing on human body image information of a plurality of different angles, then fusing the processed image information into a point cloud model, and then carrying out feature matching and cutting on the image to generate mapping information;
s3, triangularization processing and BA optimization are carried out on the generated point cloud model data with the mapping information to generate an image data set, and three-dimensional human body model reconstruction is carried out on the image data set;
s4, extracting bones of the three-dimensional human body model reconstructed in the step S3 by using a bone extraction algorithm to obtain a three-dimensional human body model bound with the bones;
s5, obtaining a human motion video stream, extracting human motion through a deep learning algorithm, adding the extracted human motion into the three-dimensional human model bound with bones in the step S4, and generating a human model with motion;
and S6, reproducing the three-dimensional human body motion.
2. The human motion three-dimensional data acquisition and reproduction processing method of claim 1, characterized in that: in S1, distortion correction processing is also required for the acquired human body image.
3. The human motion three-dimensional data acquisition and reproduction processing method of claim 2, characterized in that: in S1, the human body image is captured by a plurality of high-definition cameras.
4. A processing method for human body motion three-dimensional data acquisition and reproduction as claimed in claim 3, characterized in that: in the S1, when the plurality of high definition cameras capture the human body image, reference points of the capture terminal need to be calibrated, and the spatial orientation and the internal parameters of the capture terminal are determined.
5. The human motion three-dimensional data acquisition and reproduction processing method of claim 1, characterized in that: the bone extraction algorithm in S4 adopts a K3M sequential iteration algorithm.
6. The human motion three-dimensional data acquisition and reproduction processing method of claim 1, characterized in that: in S6, the human body model with motion generated in S5 is input to a VR reconstruction system, and the three-dimensional human body motion is reconstructed by the VR reconstruction system.
7. The human motion three-dimensional data acquisition and reproduction processing method of claim 1, characterized in that: the human body image acquired in S1 and the human body motion video stream acquired in S5 may be simultaneously obtained by different or the same photographing apparatuses.
8. A processing system for human body motion three-dimensional data acquisition and reproduction is characterized by comprising:
the image data acquisition unit is used for acquiring human body images at a plurality of different angles;
the human body model three-dimensional reconstruction unit is used for performing three-dimensional human body model reconstruction on the human body image acquired by the image data acquisition unit and extracting bones of the reconstructed three-dimensional human body model so as to obtain a three-dimensional human body model bound with the bones;
the human body action model generating unit is used for acquiring a human body motion video stream, extracting human body actions through a deep learning algorithm, adding the extracted human body actions into a three-dimensional human body model bound with bones, and generating a human body action model with actions;
the human body action three-dimensional model reproduction unit is used for reproducing the human body action after the human body action model is finished;
and the data storage unit is used for storing and interacting the data acquired by the image data acquisition unit, the data between the human body model three-dimensional reconstruction unit and the human body action model generation unit.
9. The system for processing human body motion three-dimensional data acquisition and reproduction according to claim 8, wherein: the image data acquisition unit comprises a shooting terminal consisting of a plurality of cameras capable of shooting simultaneously.
10. The system for processing human body motion three-dimensional data acquisition and reproduction according to claim 8, wherein: the human body action three-dimensional model reproduction unit comprises a VR reproduction display system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911319622.XA CN111080776B (en) | 2019-12-19 | 2019-12-19 | Human body action three-dimensional data acquisition and reproduction processing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911319622.XA CN111080776B (en) | 2019-12-19 | 2019-12-19 | Human body action three-dimensional data acquisition and reproduction processing method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111080776A true CN111080776A (en) | 2020-04-28 |
| CN111080776B CN111080776B (en) | 2023-08-29 |
Family
ID=70315868
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911319622.XA Active CN111080776B (en) | 2019-12-19 | 2019-12-19 | Human body action three-dimensional data acquisition and reproduction processing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111080776B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111882595A (en) * | 2020-07-28 | 2020-11-03 | 河海大学常州校区 | Human body semantic feature extraction method and system |
| CN113079340A (en) * | 2021-04-07 | 2021-07-06 | 西北工业大学 | Three-dimensional holographic video conference system based on Hololens2 |
| CN113487727A (en) * | 2021-07-14 | 2021-10-08 | 广西民族大学 | Three-dimensional modeling system, device and method |
| CN113823044A (en) * | 2021-10-08 | 2021-12-21 | 刘智矫 | Human body three-dimensional data acquisition room and charging method thereof |
| CN116153510A (en) * | 2023-02-17 | 2023-05-23 | 河南翔宇医疗设备股份有限公司 | Correction mirror control method, device, equipment, storage medium and intelligent correction mirror |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180158232A1 (en) * | 2016-12-07 | 2018-06-07 | Abb Schweiz Ag | Submersible inspection device and vision based modelling |
| CN109242954A (en) * | 2018-08-16 | 2019-01-18 | 叠境数字科技（上海）有限公司 | Multi-view angle three-dimensional human body reconstruction method based on template deformation |
| CN110060329A (en) * | 2019-01-15 | 2019-07-26 | 叠境数字科技（上海）有限公司 | A kind of mobile terminal manikin method for reconstructing based on color depth video stream data |
-
2019
- 2019-12-19 CN CN201911319622.XA patent/CN111080776B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180158232A1 (en) * | 2016-12-07 | 2018-06-07 | Abb Schweiz Ag | Submersible inspection device and vision based modelling |
| CN109242954A (en) * | 2018-08-16 | 2019-01-18 | 叠境数字科技（上海）有限公司 | Multi-view angle three-dimensional human body reconstruction method based on template deformation |
| CN110060329A (en) * | 2019-01-15 | 2019-07-26 | 叠境数字科技（上海）有限公司 | A kind of mobile terminal manikin method for reconstructing based on color depth video stream data |
Non-Patent Citations (1)
| Title |
|---|
| 李翔;李泽丰;李治江;: "基于Kinect的人体三维重建方法", no. 06 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111882595A (en) * | 2020-07-28 | 2020-11-03 | 河海大学常州校区 | Human body semantic feature extraction method and system |
| CN111882595B (en) * | 2020-07-28 | 2024-01-26 | 河海大学常州校区 | Human body semantic feature extraction method and system |
| CN113079340A (en) * | 2021-04-07 | 2021-07-06 | 西北工业大学 | Three-dimensional holographic video conference system based on Hololens2 |
| CN113487727A (en) * | 2021-07-14 | 2021-10-08 | 广西民族大学 | Three-dimensional modeling system, device and method |
| CN113487727B (en) * | 2021-07-14 | 2022-09-02 | 广西民族大学 | Three-dimensional modeling system, device and method |
| CN113823044A (en) * | 2021-10-08 | 2021-12-21 | 刘智矫 | Human body three-dimensional data acquisition room and charging method thereof |
| CN113823044B (en) * | 2021-10-08 | 2022-09-13 | 刘智矫 | Human body three-dimensional data acquisition room and charging method thereof |
| CN116153510A (en) * | 2023-02-17 | 2023-05-23 | 河南翔宇医疗设备股份有限公司 | Correction mirror control method, device, equipment, storage medium and intelligent correction mirror |
| CN116153510B (en) * | 2023-02-17 | 2024-04-16 | 河南翔宇医疗设备股份有限公司 | Correction mirror control method, device, equipment, storage medium and intelligent correction mirror |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111080776B (en) | 2023-08-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111080776B (en) | Human body action three-dimensional data acquisition and reproduction processing method and system | |
| CN107223269B (en) | Three-dimensional scene positioning method and device | |
| CN110998659B (en) | Image processing system, image processing method, and program | |
| JP7003994B2 (en) | Image processing equipment and methods | |
| KR100793838B1 (en) | Appratus for findinng the motion of camera, system and method for supporting augmented reality in ocean scene using the appratus | |
| US20130335535A1 (en) | Digital 3d camera using periodic illumination | |
| CN109410316B (en) | Method for three-dimensional reconstruction of object, tracking method, related device and storage medium | |
| CN109711472B (en) | Training data generation method and device | |
| US9183634B2 (en) | Image processing apparatus and image processing method | |
| CN110648274B (en) | Method and device for generating fisheye image | |
| US20210295467A1 (en) | Method for merging multiple images and post-processing of panorama | |
| CN112184811B (en) | Monocular space structured light system structure calibration method and device | |
| CN109934873B (en) | Method, device and equipment for acquiring marked image | |
| JP7421518B2 (en) | Method and apparatus for producing a three-dimensional reconstruction of an object with reduced distortion | |
| TW202145146A (en) | Image registration method and related model training methods, equipment and computer readable storage medium thereof | |
| CN115035235A (en) | Three-dimensional reconstruction method and device | |
| US11380063B2 (en) | Three-dimensional distortion display method, terminal device, and storage medium | |
| Zhang et al. | Real-time calibration and registration method for indoor scene with joint depth and color camera | |
| CN117196955A (en) | Panoramic image stitching method and terminal | |
| CN111179408B (en) | Three-dimensional modeling method and equipment | |
| CN109990756B (en) | Binocular ranging method and system | |
| JP2021128592A (en) | Image processing device, image processing method, learned model generation method, and program | |
| CN112634439B (en) | 3D information display method and device | |
| CN112652056B (en) | 3D information display method and device | |
| CN111316323A (en) | Abnormal value processing method and device for three-dimensional trajectory data |
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 |