CN111611928B - Height and body size measuring method based on monocular vision and key point identification - Google Patents
Height and body size measuring method based on monocular vision and key point identification Download PDFInfo
- Publication number
- CN111611928B CN111611928B CN202010438604.XA CN202010438604A CN111611928B CN 111611928 B CN111611928 B CN 111611928B CN 202010438604 A CN202010438604 A CN 202010438604A CN 111611928 B CN111611928 B CN 111611928B
- Authority
- CN
- China
- Prior art keywords
- height
- human body
- key
- user
- points
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/107—Measuring physical dimensions, e.g. size of the entire body or parts thereof
- A61B5/1072—Measuring physical dimensions, e.g. size of the entire body or parts thereof measuring distances on the body, e.g. measuring length, height or thickness
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/107—Measuring physical dimensions, e.g. size of the entire body or parts thereof
- A61B5/1077—Measuring of profiles
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/107—Measuring physical dimensions, e.g. size of the entire body or parts thereof
- A61B5/1079—Measuring physical dimensions, e.g. size of the entire body or parts thereof using optical or photographic means
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The invention relates to a height and body size measuring method based on monocular vision and key point identification, belongs to the field of vision measurement, and can be widely applied to body measuring equipment based on a monocular camera. According to the method, the height and the body size of the user are measured through camera image acquisition and image processing, a height measurement module based on matching of a height estimation model and a template and a body size calculation module based on key feature points of the human body. The measuring method is based on monocular vision, so that the cost of equipment can be greatly reduced; the height measurement module discards the traditional method for obtaining model parameters through physical calibration by monocular vision, and updates the parameters for different users by an iterative template matching method, so that no calibration reference object is needed to be externally added to the equipment, the equipment has stronger flexibility, and meanwhile, the parameter self-adaption and the parameter accuracy are stronger compared with the method for fixing the parameters.
Description
Technical Field
The invention belongs to the field of vision measurement, and particularly relates to a height and body size measurement method based on monocular vision and key point identification.
Background
In order to acquire the height information of a human body, a depth camera is often adopted by the present measuring equipment to acquire the depth information of a human body image. However, depth cameras are expensive, so that monocular cameras are an option for volume measuring devices in order to reduce costs. The measuring model parameters of the measuring body equipment based on the monocular camera often need to be calibrated by a reference object, so that the flexibility of the equipment is limited to a certain extent; the calibrated parameters are fixed parameters, so that the method has no robustness. Under the background, the invention realizes the self-adaptive change of the model parameters in a key feature point template matching mode, reduces the cost by using the monocular camera, enhances the robustness of the model, and is generally suitable for low-cost and high-robustness measuring equipment.
Disclosure of Invention
The invention aims to provide a height and body size measuring method based on monocular vision and key point identification, which can reduce the cost of measuring equipment and improve the robustness of the measuring result of the equipment.
The invention adopts the following measurement technical scheme:
a height and body size measuring method based on monocular vision and key point identification mainly comprises the following steps:
s1, a data acquisition unit: when the gesture of the user is detected to be correct, the equipment camera automatically acquires the whole body picture of the user;
s2, an image processing unit: extracting edge contour points of a human body by combining a semantic segmentation model, detecting and identifying key nodes of the human body through key nodes of the human body, and further expanding key feature points of the human body according to the contour points and the key nodes;
s3, height calculating unit: the parameters of the height calculation model of the user are obtained through matching between the user and the template and self-adaptive updating, and the real height of the user is calculated through the height calculation model after parameter updating;
s4, body size calculating unit: and (3) calculating the data of each size by combining the key characteristic points of the user obtained in the step (S2) and the height information obtained in the step (S3).
Further, the following are included in step S1:
1) Detecting human bodies in real time, and detecting anchor point frame areas of the human bodies;
2) Detecting human body key points in real time, wherein the human body key points comprise arms, legs, heads and trunk parts;
3) Judging whether the gesture of the user is correct according to the distribution condition of the key points of the human body in the anchor point frame. Human body standard posture: the front part is visually observed, and the body is opposite to the camera; the double arms are unfolded for 30-45 degrees, and the double legs are separated to have the same width as the shoulders;
the gesture of the user is automatically grasped, and no additional manual operation is needed.
Further, the step S2 further includes the following:
1) Firstly, carrying out sparse processing and smooth noise reduction processing on coordinates of human body edge contour points, wherein the sparse processing is to adaptively select sparse coefficients according to areas of human body areas in images;
2) According to the structural characteristics of body parts such as shoulders, chest, waist, middle waist and buttocks of a human body and the detected key node coordinates of the human body, the vertex coordinates of the human body and other key characteristic points related to the body size, such as waist edge points, buttock edge points and the like, are found in the processed edge contour points.
Further, the step S3 further includes the following:
1) Collecting sample data: in order to avoid the traditional calibration mode, the invention firstly collects M groups (M reference value is more than 10) of real person characteristic points and height data as a sample data set, and initializes the height calculation model parameters through the sample data without externally connecting an auxiliary calibration reference object. The sample is not required to be recorded again after being collected once;
2) Height calculation model and parameter acquisition: height calculation model: the precondition of the height calculation model is that the plane of the camera lens on the equipment is vertical to the ground. In the image coordinate system, the ordinate of the top of the human body in the image is t, the ordinate of the key point of the human heel is b, the height s=b-t of the human body in the image coordinate system, the ordinate of the vanishing line in the image coordinate system is a, and the height of the vanishing line from the human heel in the image coordinate system is l=b-a; in the world coordinate system, the user' S real height is S, the vanishing line (virtual line, not actually present) in the world coordinate system is L (the reference height may be the lens height from the ground), there is S/l=s/L according to the perspective principle, i.eWherein L, a is a super parameter, b, t is known (can be obtained according to key feature point coordinates in an image), and the real height S is a target value; parameter acquisition: collecting a certain amount of real human body height and picture sample data, updating the super parameters L through a regression model, and a completing the parameter updating of a one-time height calculation model, and then, obtaining the real human body height and picture sample data by using the regression modelDirectly calculating to obtain the real height S of the human body;
3) Initializing a template and initializing parameters of a height calculation model: template matching is carried out on the key feature points of the user and the key feature points of the sample set, so that an array sample template subset with higher matching degree is obtained, namely an initial template sample; initializing parameters of a height calculation model through an initial template sample;
4) Iterative updating of height calculation model parameters:
after calculating model parameters by updating the height each time, estimating the height of the user through the model and matching with the template, finding an array sample with higher matching degree as an updated template sample, updating the model parameters according to the template sample, and iterating the process to n times, wherein the reference value is n=3.
Drawings
FIG. 1 is a schematic illustration of the application of a height and body size measurement method based on monocular vision and key point recognition to a body measurement device equipped with a monocular camera;
FIG. 2 is a flow chart of height measurement and body size calculation.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which embodiments of the invention are shown, it being evident that the embodiments described are only some, but not all, of the embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Embodiment 1, as shown in fig. 1 and 2, a height and body size measuring method based on monocular vision and key point recognition comprises the following steps;
the first step: sample data set entry: the data preparation firstly records sample data (including key feature point information and height information of human body) of a real human body on the equipment as a to-be-determined matching template. The advantage of collecting samples in advance: instead of the traditional parameter acquisition mode of a fixed reference object, the parameter acquisition method can adaptively generate the height measurement model parameters suitable for users through different template sets, and improves the adaptability of the model parameters. And the user does not need to enter again after entering once.
And a second step of: user picture input: when the actual user measures, the user stands according to the required gesture, and the camera automatically judges the correctness of the gesture according to the distribution of key feature points in the human body anchor point frame and captures a correct gesture image. Human body standard posture: the front part is visually observed, and the body is opposite to the camera; the arms are unfolded for 30-45 degrees, and the legs are separated by the feet and the shoulders to be the same width. The mode of front-end human body detection has the following advantages: the image acquisition process is automatically completed without additional manual operation.
And a third step of: calculating the height of a user: and matching the user with the template, adaptively updating to obtain parameters of the height calculation model of the user, and calculating the real height of the user through the height calculation model after parameter updating. The method has the advantages that: the fixed model parameters are banned, the height model parameters aiming at different users are obtained in a self-adaptive mode, and the robustness and the accuracy are high.
Fourth step: the method comprises the steps of firstly, carrying out sparse processing and smooth noise reduction processing on human body edge contour point coordinates by a computing body, wherein the sparse processing is to adaptively select sparse coefficients according to the area of a human body region in an image; according to the structural characteristics of body parts such as shoulders, chest, waist, middle waist and buttocks of a human body and the detected key node coordinates of the human body, the vertex coordinates of the human body and other key characteristic points related to the body size, such as waist edge points, buttock edge points and the like, are found in the processed edge contour points. The human body contour sparsity adopts a self-adaptive sparsity method, and proper sparsity coefficients are selected in a self-adaptive mode according to the size of the region of the human body, so that the advantages are that: the reasonable number of the contour points is ensured according to the distance between the human body and the lens, the calculated amount when the key feature points are searched in the contour points is saved, and meanwhile, the precision of the key feature points is ensured.
Fifth step: and calculating the body size of the user according to the key feature points and the height information of the user.
Claims (3)
1. The utility model provides a height and body size measuring method based on monocular vision and key point discernment, characterized by that, measuring method includes following unit:
s1, a data acquisition unit: when the gesture of the user is detected to be correct, the equipment camera automatically acquires the whole body picture of the user;
s2, an image processing unit: extracting edge contour points of a human body by combining a semantic segmentation model, detecting and identifying key nodes of the human body through key nodes of the human body, and further expanding key feature points of the human body according to the contour points and the key nodes;
s3, height calculating unit: the parameters of the height calculation model of the user are obtained through matching between the user and the template and self-adaptive updating, and the real height of the user is calculated through the height calculation model after parameter updating;
s4, body size calculating unit: combining the key feature points of the user obtained in the step S2 and the height information obtained in the step S3, and calculating the data of each size;
the specific steps of the height calculating unit in the step S3 are as follows:
1) Collecting sample data: in order to avoid the traditional calibration mode, firstly, M groups of true human feature points and height data are collected as sample data sets, M reference values are more than 10, height calculation model parameters are initialized through the sample data, an external auxiliary calibration reference object is not needed, and re-input is not needed after one sample collection is achieved;
2) Height calculation model and parameter acquisition: the method comprises the following steps of a height calculation model, wherein the precondition of the height calculation model is that a plane of a camera lens on equipment is perpendicular to the ground, in an image coordinate system, the ordinate of the top of a human body is t, the ordinate of a key point of a human heel is b, the height s=b-t of the human body in the image coordinate system, the ordinate of a vanishing line in the image coordinate system is a, and the height of the vanishing line from the human heel in the image coordinate system is l=b-a; in the world coordinate system, the real height of the user is S, and the height of the vanishing line in the world coordinate system is L; according to perspective principle there is S/l=s/L, i.eWherein L, a is a super parameter, b, t is known, and the real height S is a target value; parameter acquisition: collecting a certain amount of real human body height and picture sample data, updating super parameters L through a regression model, a completing parameter updating of a one-time height calculation model, and then enabling the user to use +.>Directly calculating to obtain the real height S of the human body;
3) Initializing a template and initializing parameters of a height calculation model: template matching is carried out on the key feature points of the user and the key feature points of the sample set, so that an array sample template subset with higher matching degree is obtained, namely an initial template sample; initializing parameters of a height calculation model through an initial template sample;
4) Iterative updating of height calculation model parameters:
after calculating model parameters by updating the height each time, estimating the height of the user through the model and matching with the template, finding an array sample with higher matching degree as an updated template sample, updating the model parameters according to the template sample, and iterating the process for n times, wherein n=3.
2. The method for measuring height and body dimensions based on monocular vision and keypoint recognition according to claim 1, wherein the specific steps of detecting the user gesture in step S1 are as follows:
1) Detecting human bodies in real time, and detecting anchor point frame areas of the human bodies;
2) Detecting human body key points in real time, wherein the human body key points comprise arms, legs, heads and trunk parts;
3) Judging whether the gesture of the user is correct according to the distribution condition of the key points of the human body in the anchor point frame, and judging whether the gesture of the human body is standard gesture: the front part is visually observed, and the body is opposite to the camera; the arms are unfolded for 30-45 degrees, and the legs are separated by the feet and the shoulders to be the same width.
3. The method for measuring height and body size based on monocular vision and key point recognition according to claim 1, wherein the step of expanding key feature points of the human body in the step S2 is as follows,
1) Firstly, carrying out sparse processing and smooth noise reduction processing on coordinates of human body edge contour points, wherein the sparse processing is to adaptively select sparse coefficients according to areas of human body areas in images;
2) According to the structural characteristics of the body parts of the shoulders, the chest, the waist, the middle waist and the buttocks of the human body and the detected key node coordinates of the human body, the vertex coordinates of the human body and other key characteristic points related to the body size are found out in the processed edge contour points, and the other key characteristic points related to the body size comprise waist edge points and buttock edge points.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010438604.XA CN111611928B (en) | 2020-05-22 | 2020-05-22 | Height and body size measuring method based on monocular vision and key point identification |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010438604.XA CN111611928B (en) | 2020-05-22 | 2020-05-22 | Height and body size measuring method based on monocular vision and key point identification |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111611928A CN111611928A (en) | 2020-09-01 |
CN111611928B true CN111611928B (en) | 2023-07-28 |
Family
ID=72199472
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010438604.XA Active CN111611928B (en) | 2020-05-22 | 2020-05-22 | Height and body size measuring method based on monocular vision and key point identification |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111611928B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111930240B (en) * | 2020-09-17 | 2021-02-09 | 平安国际智慧城市科技股份有限公司 | Motion video acquisition method and device based on AR interaction, electronic equipment and medium |
CN113763453A (en) * | 2021-09-06 | 2021-12-07 | 北京云数工场数据科技有限公司 | Artificial intelligence energy system and method |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105930581A (en) * | 2016-04-20 | 2016-09-07 | 北京长城华冠汽车科技股份有限公司 | Method and device for determining vertical distance between steering wheel center and heel point |
CN105956539A (en) * | 2016-04-27 | 2016-09-21 | 武汉大学 | Method for height measurement of human body based on background modeling and binocular vision |
CN108596888A (en) * | 2018-04-19 | 2018-09-28 | 李富平 | A kind of Human Height Real-time Generation based on monocular image |
JP2018195050A (en) * | 2017-05-17 | 2018-12-06 | 株式会社デンソーアイティーラボラトリ | Object detecting device, object detecting method, and program |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2893479B1 (en) * | 2012-09-05 | 2018-10-24 | Sizer Technologies Ltd | System and method for deriving accurate body size measures from a sequence of 2d images |
PT2757346E (en) * | 2013-01-21 | 2016-06-02 | Kapsch Trafficcom Ag | Method for measuring a height profile of a vehicle passing on the street |
CN107507213A (en) * | 2016-06-13 | 2017-12-22 | 北京理工大学 | A kind of method and device based on end point extraction image information |
CN107041585A (en) * | 2017-03-07 | 2017-08-15 | 上海优裁信息技术有限公司 | The measuring method of human dimension |
US10832472B2 (en) * | 2018-10-22 | 2020-11-10 | The Hong Kong Polytechnic University | Method and/or system for reconstructing from images a personalized 3D human body model and thereof |
CN109938737A (en) * | 2019-03-01 | 2019-06-28 | 苏州博慧智能科技有限公司 | A kind of human body body type measurement method and device based on deep learning critical point detection |
CN110263662B (en) * | 2019-05-28 | 2021-05-18 | 北京智形天下科技有限责任公司 | Human body contour key point and key part identification method based on grading |
-
2020
- 2020-05-22 CN CN202010438604.XA patent/CN111611928B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105930581A (en) * | 2016-04-20 | 2016-09-07 | 北京长城华冠汽车科技股份有限公司 | Method and device for determining vertical distance between steering wheel center and heel point |
CN105956539A (en) * | 2016-04-27 | 2016-09-21 | 武汉大学 | Method for height measurement of human body based on background modeling and binocular vision |
JP2018195050A (en) * | 2017-05-17 | 2018-12-06 | 株式会社デンソーアイティーラボラトリ | Object detecting device, object detecting method, and program |
CN108596888A (en) * | 2018-04-19 | 2018-09-28 | 李富平 | A kind of Human Height Real-time Generation based on monocular image |
Non-Patent Citations (2)
Title |
---|
FITME: BODY MEASUREMENT ESTIMATIONS USING MACHINE LEARNING METHOD;Sahar Ashmawi.et.;《Procedia Computer Science 》;第163卷;第209-217页 * |
基于红外图像的人体参数测量技术研究;杨泷迪;《中国优秀硕士学位论文全文数据库信息科技辑》(第3期);第I138-1353页 * |
Also Published As
Publication number | Publication date |
---|---|
CN111611928A (en) | 2020-09-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108765414B (en) | No-reference stereo image quality evaluation method based on wavelet decomposition and natural scene statistics | |
CN111611928B (en) | Height and body size measuring method based on monocular vision and key point identification | |
CN111914642B (en) | Pedestrian re-identification method, device, equipment and medium | |
CN109725721B (en) | Human eye positioning method and system for naked eye 3D display system | |
CN108960142B (en) | Pedestrian re-identification method based on global feature loss function | |
CN107767358B (en) | Method and device for determining ambiguity of object in image | |
CN113052185A (en) | Small sample target detection method based on fast R-CNN | |
CN105279522A (en) | Scene object real-time registering method based on SIFT | |
CN107292299A (en) | Side face identification method based on kernel specification correlation analysis | |
CN116071424A (en) | Fruit space coordinate positioning method based on monocular vision | |
CN110188684B (en) | Face recognition device and method | |
CN115908995A (en) | Digital instrument reading identification method and device, electronic equipment and storage medium | |
CN106886754B (en) | Object identification method and system under a kind of three-dimensional scenic based on tri patch | |
CN108710841A (en) | A kind of face living body detection device and method based on MEMs infrared sensor arrays | |
CN105654479A (en) | Multispectral image registering method and multispectral image registering device | |
WO2022170896A9 (en) | Key point detection method and system, intelligent terminal, and storage medium | |
CN108447084B (en) | Stereo matching compensation method based on ORB characteristics | |
CN106971176A (en) | Tracking infrared human body target method based on rarefaction representation | |
CN102201060B (en) | Method for tracking and evaluating nonparametric outline based on shape semanteme | |
CN111325282A (en) | Mammary gland X-ray image identification method and device suitable for multiple models | |
CN116612497A (en) | Clothing changing pedestrian re-identification method based on clothing style feature fusion | |
CN113850769B (en) | Hyperspectral change detection method based on Simese space spectrum joint convolution network | |
CN117152787A (en) | Character clothing recognition method, device, equipment and readable storage medium | |
KR101842797B1 (en) | Apparatus and method for analyzing 3-dimensional body pose | |
Liu et al. | Multi-view ear shape feature extraction and reconstruction |
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 | ||
TA01 | Transfer of patent application right | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20230703 Address after: 2-1-1, No. 5-533, Langyun Street, Hunnan District, Shenyang, Liaoning 110623 Applicant after: Zheng Zeyu Address before: Room 101-4, building 1, Juhua building, 6 Longzhou Road, Cangqian street, Yuhang District, Hangzhou City, Zhejiang Province 310000 Applicant before: Hangzhou Zhiron Intelligent Technology Co.,Ltd. |
|
GR01 | Patent grant | ||
GR01 | Patent grant |