WO2021135138A1 - 一种目标运动轨迹构建方法、设备以及计算机存储介质 - Google Patents
一种目标运动轨迹构建方法、设备以及计算机存储介质 Download PDFInfo
- Publication number
- WO2021135138A1 WO2021135138A1 PCT/CN2020/100265 CN2020100265W WO2021135138A1 WO 2021135138 A1 WO2021135138 A1 WO 2021135138A1 CN 2020100265 W CN2020100265 W CN 2020100265W WO 2021135138 A1 WO2021135138 A1 WO 2021135138A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- target
- picture
- feature
- features
- face
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/75—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/783—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/787—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using geographical or spatial information, e.g. location
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- 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
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- 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
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
- G06V40/173—Classification, e.g. identification face re-identification, e.g. recognising unknown faces across different face tracks
-
- 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
-
- 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
- G06T2207/30232—Surveillance
-
- 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
- G06T2207/30236—Traffic on road, railway or crossing
-
- 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
- G06T2207/30241—Trajectory
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/08—Detecting or categorising vehicles
Definitions
- This application relates to the field of traffic monitoring, and in particular to a method, equipment and computer storage medium for constructing a target motion trajectory.
- This application provides a method, equipment and computer-readable storage medium for constructing a target motion trajectory.
- the present application provides a method for constructing a target motion trajectory.
- the method for constructing a target motion trajectory includes:
- At least two different types of target features that match the retrieval condition, where the at least two different types of target features include at least two of face features, human body features, and vehicle features;
- the target motion trajectory is generated according to the combination of the shooting time and the shooting location associated with the at least two different types of target features.
- the step of generating the target motion trajectory according to the combination of the shooting time and the shooting location associated with the at least two different types of target features further includes:
- the shooting time and shooting location associated with the auxiliary target feature are eliminated.
- the method of judging whether the relative position of the auxiliary target feature and the main target feature conforms to the motion law of the target according to the shooting time and shooting location of the main target feature, and the shooting time and shooting location of the auxiliary target feature further include:
- the movement speed is calculated based on the position difference and the time difference, and when the movement speed is less than or equal to a preset movement speed threshold, it is determined that the relative position of the auxiliary target feature and the main target feature conforms to the movement law of the target.
- the acquiring the shooting time and the shooting location respectively associated with the at least two different types of target features includes:
- the shooting time and shooting location associated with the target feature are determined based on at least the first target picture.
- the method further includes:
- a target human face picture corresponding to the human face feature Acquiring a target human face picture corresponding to the human face feature, a target human body picture corresponding to the human body feature, and/or a target vehicle picture corresponding to the vehicle feature;
- the target human face picture and the target human body picture correspond to the same first target picture and have a preset spatial relationship
- the target human face picture and the target human body in the first target picture Picture association in the case where the target face picture and the target vehicle picture correspond to the same first target picture and have a preset spatial relationship, combine the target face picture in the first target picture with all The target vehicle picture is associated; in the case that the target human body picture and the target vehicle picture correspond to the same first target picture and have a preset spatial relationship, combine the target human body picture in the first target picture with The target vehicle picture is associated.
- the Methods also include:
- the determining the shooting time and shooting location associated with the target feature based at least on the first target picture includes:
- the shooting time and shooting location associated with the target feature are determined based on the first target picture and the second target picture.
- the Methods also include:
- the determining the shooting time and shooting location associated with the target feature based at least on the first target picture includes:
- the shooting time and shooting location associated with the target feature are determined based on the first target picture and the third target picture.
- the preset spatial relationship includes at least one of the following:
- the image coverage of the first target-related picture includes the image coverage of the second target-related picture
- the image coverage of the first target-related picture partially overlaps the image coverage of the second target-related picture
- the image coverage of the first target-related picture is connected to the image coverage of the second target-related picture
- the first target associated picture includes any one or more of the target face picture, the target human body picture, and the target vehicle picture
- the second target associated picture includes the target person Any one or more of the face picture, the target human body picture, and the target vehicle picture.
- the step of obtaining at least two different types of target features that match the retrieval conditions includes:
- the target feature that matches any one of the at least two search conditions is retrieved from the database.
- the retrieval conditions include at least one of identity retrieval conditions, face retrieval conditions, human body retrieval conditions, and vehicle retrieval conditions;
- the target feature is pre-associated with identity information
- the identity information is any one of ID card information, name information, and file information.
- the step of retrieving the target feature matching any one of the at least two retrieval conditions from the database includes:
- the present application provides a target motion trajectory construction device.
- the target motion trajectory construction device includes a processor and a memory.
- a computer program is stored in the memory.
- the processor is used to execute the computer program to implement the steps of the target motion trajectory construction method.
- the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed, the steps of constructing the above-mentioned target motion trajectory are realized.
- This application also provides a computer program product.
- the instructions in the computer program product are executed by a processor, any one of the aforementioned methods for constructing a target motion trajectory is executed.
- the beneficial effect of the present application is that the target motion trajectory construction device acquires at least two different types of target features that match the retrieval conditions, wherein the at least two different types of target features include at least one of a face feature, a human body feature, and a vehicle feature.
- Two types acquiring shooting time and shooting location respectively associated with at least two different types of target features; generating a target motion trajectory according to a combination of shooting time and shooting location associated with at least two different types of target features.
- Fig. 1 is a schematic flowchart of a first embodiment of a method for constructing a target motion trajectory provided by the present application
- FIG. 2 is a schematic flowchart of a second embodiment of a method for constructing a target motion trajectory provided by the present application
- FIG. 3 is a schematic flowchart of a third embodiment of a method for constructing a target motion trajectory provided by the present application
- FIG. 4 is a schematic flowchart of a fourth embodiment of a method for constructing a target motion trajectory provided by the present application
- Fig. 5 is a schematic structural diagram of an embodiment of a target motion trajectory construction device provided by the present application.
- Fig. 6 is a schematic structural diagram of another embodiment of a target motion trajectory construction device provided by the present application.
- Fig. 7 is a schematic structural diagram of an embodiment of a computer-readable storage medium provided by the present application.
- This application provides a specific method for constructing a target motion trajectory. Based on the development of face retrieval, human body retrieval, vehicle retrieval and video structuring technology, the method provided by this application integrates multiple algorithms to automatically combine single retrieval objects or multiple retrieval objects such as face information, human body information, and vehicle information in traffic images. The combination of the retrieval objects retrieves the results at the same time at one time, and merges and restores all the target motion trajectories.
- FIG. 1 is a schematic flowchart of a first embodiment of a method for constructing a target motion trajectory provided by the present application.
- the target motion trajectory construction method of the present application is applied to a target motion trajectory construction device.
- the target motion trajectory construction device can be a terminal device such as a smart phone, a tablet computer, a notebook computer, a computer or a wearable device, or it can be a bayonet traffic The monitoring system in the system.
- a trajectory construction device is used uniformly to describe the method for constructing a target motion trajectory.
- the method for constructing a target motion trajectory of this embodiment specifically includes the following steps:
- S101 Acquire at least two different types of target features that match the retrieval condition, where the at least two different types of target features include at least two of face features, human body features, and vehicle features.
- the trajectory construction equipment obtains multiple image data, and the image data can be obtained directly from the existing traffic big data open source platform or from the traffic management department.
- the image data includes time information and location information.
- the trajectory construction equipment can also obtain a real-time video stream from an existing traffic big data open source platform or from a traffic management department, and then perform image frame segmentation on the real-time video stream to obtain multiple image data.
- the image data needs to include the bayonet point position information in the monitoring area, such as latitude and longitude (latitude, longitude) information, etc.; it also needs to include a preset time period, such as the bayonet snapped car record data within a month, where , Bayonet captures the past car record data including time information. If location information such as latitude and longitude is stored in the bayonet-captured vehicle history data, the bayonet point position information can also be directly extracted from the bayonet-capture vehicle history data.
- the capture records in the recent period of time cannot guarantee that all the bayonet points have image data.
- the terminal equipment needs to be upgraded from the existing traffic.
- the data open source platform or the traffic management department obtains all the bayonet point location information.
- the terminal device can also preprocess the image data. Specifically, the terminal device determines whether each image data includes time information of the capture time and all information in the location information including latitude and longitude information. If any one of the time information and the location information is missing in the image data, the terminal device directly removes the corresponding image data, so as to avoid the problem of data missing in the subsequent spatio-temporal prediction database.
- the terminal equipment cleans the repeated data and invalid data in the original image data, which is conducive to data analysis.
- the trajectory construction device performs target detection on multiple image data respectively. Specifically, the trajectory construction device detects all the faces, human bodies and/or vehicles in the image data through a target detection algorithm or the fusion of multiple target detection algorithms. , And extract all the features of faces, human bodies and/or vehicles to form target features.
- the target feature may include image features extracted from image data and/or text features generated by structural analysis of image features.
- Image features include all facial features, human body features, and vehicle features in image data.
- Text features are feature information generated by structural analysis of vehicle features. For example, trajectory building equipment can perform text recognition on vehicle features to obtain vehicle features.
- the license plate number in, the license plate number is used as a text feature.
- the trajectory construction device receives the search condition input by the user, and searches the dynamic database for the target feature matching the search condition according to the search condition.
- the trajectory construction device acquires at least two different types of target features that match the retrieval conditions, and the at least two different types of target features include at least two of face features, human body features, and vehicle features. Obtaining multiple types of target features is conducive to extracting sufficient trajectory information, avoiding the loss of some important trajectory information due to blurred shooting, obstructions, etc., and improving the accuracy of trajectory construction methods.
- the retrieval conditions can be the human face and human body images of the retrieval target obtained by the police through on-site surveys, police station reporting, and capture retrieval, etc., or any images or texts containing the above-mentioned image information.
- the trajectory construction device retrieves the target feature matching the face and human body image from the dynamic database according to the face and human body image.
- S102 Acquire shooting time and shooting location respectively associated with at least two different types of target features.
- the trajectory construction device after acquiring the target feature of the image data, the trajectory construction device further acquires the shooting time and shooting location of the image data, and associates the target feature of the same image data with the corresponding shooting time and shooting location.
- the association method can be stored in the same storage space, or the same identification number can be set.
- the trajectory construction device obtains the shooting time of the target feature from the time information of the image data, and the trajectory construction device obtains the shooting location of the target feature from the position information of the image data.
- the trajectory construction device further stores the associated target feature, shooting time and shooting location in a dynamic database, where the dynamic database settings can be in the server, in the local storage, or in the cloud.
- S103 Generate a target motion trajectory according to a combination of shooting time and shooting location associated with at least two different types of target features.
- the trajectory construction device extracts the shooting time and shooting location associated with the target feature matching the retrieval condition from the dynamic database, and connects the shooting locations according to the sequence of the target feature, that is, the shooting time sequence, to generate the target motion trajectory.
- the target motion trajectory construction device acquires at least two different types of target features that match the retrieval conditions, and the at least two different types of target features include at least two of face features, human body features, and vehicle features.
- the at least two different types of target features include at least two of face features, human body features, and vehicle features.
- FIG. 2 is a view of the second embodiment of the target motion trajectory construction method provided by this application. Schematic diagram of the process.
- the method for constructing a target motion trajectory of this embodiment specifically includes the following steps:
- S201 Acquire at least two retrieval conditions.
- the at least two search conditions shown in this application include at least two of a face search condition, a human body search condition, and a vehicle search condition. Based on the types of search conditions mentioned above, this application also provides corresponding search methods.
- the trajectory construction device acquires an image data and uses any target or combination of targets such as a face, a human body, and a vehicle as the retrieval condition
- the types of retrieval algorithms automatically invoked by the trajectory construction device are:
- the search condition may also include an identity search condition, wherein the above-mentioned target feature is pre-associated with identity information, and the identity information is any one of ID card information, name information, and file information.
- S202 Search the database for the target feature that matches any one of the at least two search conditions.
- the trajectory construction device retrieves the required target features in the dynamic database, it respectively matches the target features with at least two retrieval conditions input by the user, and selects the target that matches any one of the at least two retrieval conditions. feature.
- the trajectory construction device searches in the dynamic database based on the face search condition and the vehicle search condition, and extracts the search condition and the vehicle search condition. At least one of the search conditions matches the target feature, thereby realizing the multi-dimensional retrieval of the target feature and avoiding the problem of missing track points caused by single-dimensional retrieval.
- the face search method based on the face search condition is specifically: the face in the image uploaded by the user is compared with the face of the target feature in the dynamic database, and the target feature whose similarity exceeds a set threshold is returned.
- the fusion retrieval method based on the face retrieval condition and the human body retrieval condition is specifically: the face or human body in the image uploaded by the user is compared with the face or human body of the target feature in the dynamic database, and the similarity exceeds the set threshold.
- Target characteristics are specifically: the face in the image uploaded by the user is compared with the face of the target feature in the dynamic database, and the similarity exceeds the set threshold.
- the vehicle retrieval method based on vehicle retrieval conditions is specifically: the vehicle in the image uploaded by the user is compared with the vehicle of the target feature in the dynamic database, and the target feature whose similarity exceeds the set threshold is returned; the vehicle retrieval method can also be input by the user
- the license plate number is searched for the license plate number extracted in the dynamic database structured, and the target feature corresponding to the license plate number is returned.
- the face retrieval method based on the face retrieval condition is specifically as follows: the user inputs any one of ID card information, name information, and file information, and the target feature corresponding to the identity information is matched and associated based on the above information. For example, when the police needs to hunt down a suspect, they can input the identity information of the suspect into the trajectory construction device.
- the identity information can be any of file ID, name, ID card, and license plate number.
- the trajectory construction device uses the sample feature of any one of the at least two retrieval conditions input by the user as the clustering center, clusters the target features in the database, and classifies the target within the preset range of the clustering center
- the feature is used as the target feature that matches the search condition.
- the trajectory construction setting retrieves the target feature through any two retrieval conditions of the face retrieval condition, the human body retrieval condition, the vehicle retrieval condition, and the identity retrieval condition, which can realize multi-dimensional retrieval and improve the accuracy of retrieval. And efficiency.
- the present application also provides another specific target motion trajectory construction method.
- FIG. 3 is the flow of the third embodiment of the target motion trajectory construction method provided by the present application. Schematic.
- the method for constructing a target motion trajectory of this embodiment specifically includes the following steps:
- S301 Use a target feature of at least two different types of target features as a main target feature, and use other types of target features as auxiliary target features.
- the trajectory construction device sets face features as the main target feature, and other types of target features, such as human body features and vehicle features, as secondary target features. .
- S302 Determine whether the relative position of the auxiliary target feature and the main target feature conforms to the motion law of the target according to the shooting time and shooting location of the main target feature, and the shooting time and shooting location of the auxiliary target feature.
- the trajectory construction device acquires adjacent main target features and auxiliary target features, calculates the displacement difference according to the shooting location of the main target feature and the shooting location of the auxiliary target feature, and the shooting time according to the shooting time of the main target feature and the auxiliary target feature Time calculation time difference. Furthermore, the trajectory construction device calculates the movement speed between the main target feature and the auxiliary target feature based on the displacement difference and the time difference.
- the trajectory construction device can preset a movement speed threshold based on the maximum speed limit of the road, interval speed measurement data, historical pedestrian data, and the like.
- a movement speed threshold based on the maximum speed limit of the road, interval speed measurement data, historical pedestrian data, and the like.
- the trajectory construction device detects the relationship between the target features and judges whether it conforms to the motion law of the target, thereby eliminating the shooting time and shooting location associated with the wrong target feature, thereby improving the accuracy of the target motion trajectory building method .
- FIG. 4 is a flowchart of the fourth embodiment of the target motion trajectory construction method provided by this application. Schematic.
- the method for constructing a target motion trajectory of this embodiment specifically includes the following steps:
- S401 Acquire first target pictures respectively corresponding to at least two different types of target features.
- the trajectory construction device obtains a first target picture, and the first target picture includes at least two different types of target features.
- the trajectory construction device separately obtains the target face picture corresponding to the face feature, the target human picture corresponding to the human body feature, and the target vehicle picture corresponding to the vehicle feature.
- the pictures may exist in the same first target picture.
- the trajectory construction device When the target face picture, target body picture, and/or target vehicle picture exist in the same first target picture, the trajectory construction device further associates the target face picture, target body picture, and/or target vehicle picture according to the preset spatial relationship .
- the preset spatial relationship includes any one of the following: the image coverage of the target vehicle picture includes the image coverage of the target face picture; the image coverage of the target vehicle picture It partially overlaps with the image coverage of the target face picture; the image coverage of the target vehicle picture is connected with the image coverage of the target face picture.
- the predetermined spatial relationship is used to determine whether there is an association relationship between the target face picture, the target human body picture, and the target vehicle picture, which can quickly and accurately identify the relationship between the face, the human body, and the vehicle.
- the coverage of the target vehicle picture includes the coverage of the target face picture of the driver inside the vehicle. Therefore, it is judged that the two are related to each other and are related to each other; In this case, the image coverage of the target human body picture of the cyclist partially overlaps the image coverage of the target vehicle picture. Therefore, it is determined that the two have an associated relationship and are associated with each other.
- the trajectory building device obtains the target vehicle picture corresponding to the target vehicle picture based on the target vehicle picture.
- Obtaining the second target picture corresponding to the target vehicle picture and the third target picture corresponding to the target human body picture is for when a target picture does not include the target face image, it can be based on the association relationship and the target vehicle picture and/or The target human body image searches for the target face image to enrich the trajectory information constructed by the target motion trajectory.
- S402 Determine the shooting time and shooting location associated with the target feature based at least on the first target picture.
- the trajectory construction device determines the shooting time and shooting location associated with the target feature based on the first target picture, the second target picture, and/or the third target picture.
- this application also provides a target motion trajectory construction device.
- FIG. 5 is a schematic structural diagram of an embodiment of the target motion trajectory construction device provided by this application.
- the target motion trajectory construction device 500 of this embodiment can be used to execute or implement the target motion trajectory construction method in any of the above embodiments. As shown in FIG. 5, the target motion trajectory construction device 500 includes a retrieval module 51, an acquisition module 52 and a trajectory construction module 53.
- the retrieval module 51 is configured to obtain at least two different types of target features matching the retrieval conditions, wherein the at least two different types of target features include at least two of face features, human body features, and vehicle features.
- the acquiring module 52 is configured to acquire the shooting time and shooting location respectively associated with at least two different types of target features.
- the trajectory construction module 53 is configured to generate a target motion trajectory according to the combination of the shooting time and the shooting location associated with the at least two different types of target features.
- the trajectory construction module 53 is further configured to use a target feature of a certain type among the at least two different types of target features as the main target feature, and use other types of target features as the auxiliary target feature. According to the shooting time and shooting location of the main target feature, and the shooting time and shooting location of the auxiliary target feature, it is judged whether the relative position of the auxiliary target feature and the main target feature conforms to the motion law of the target. If it does not conform to the movement law of the target, the shooting time and shooting location associated with the auxiliary target feature are eliminated.
- the trajectory construction module 53 is further configured to: calculate the position difference according to the shooting location of the main target feature and the shooting location of the auxiliary target feature; calculate the time difference according to the shooting time of the main target feature and the shooting time of the auxiliary target feature; The position difference and the time difference calculate the movement speed. When the movement speed is greater than the preset movement speed threshold, it is judged that the relative position of the auxiliary target feature and the main target feature does not conform to the movement law of the target.
- the acquiring module 52 is further configured to: acquire first target pictures respectively corresponding to at least two different types of target features; and determine the shooting time and shooting location associated with the target features based at least on the first target picture.
- the acquiring module 52 is also used to: respectively acquire the target face picture corresponding to the face feature, the target human body picture corresponding to the human body feature, and/or the target vehicle picture corresponding to the vehicle feature; When the human body picture corresponds to the same first target picture and has a preset spatial relationship, associate the target face picture in the first target picture with the target human body picture; when the target face picture and the target vehicle picture correspond to the same first target picture When the target picture has a preset spatial relationship, associate the target face picture in the first target picture with the target vehicle picture; when the target human body picture and the target vehicle picture correspond to the same first target picture and have the preset spatial relationship In the case of, the target human body picture in the first target picture is associated with the target vehicle picture.
- the acquiring module 52 is further configured to: The target vehicle picture acquires a second target picture corresponding to the target vehicle picture; and the shooting time and shooting location associated with the target feature are determined based on the first target picture and the second target picture.
- the acquiring module 52 is further configured to: The target human body picture acquires a third target picture corresponding to the target human body picture; and the shooting time and shooting location associated with the target feature are determined based on the first target picture and the third target picture.
- the preset spatial relationship includes at least one of the following: the image coverage of the first target associated picture includes the image coverage of the second target associated picture; the image coverage of the first target associated picture is associated with the second target The image coverage areas of the pictures partially overlap; the image coverage areas of the first target-associated pictures are connected with the image coverage areas of the second target-associated pictures.
- the first target-related picture includes any one or more of the target face picture, the target human body picture, and the target vehicle picture
- the second target-related picture includes any one of the target face picture, the target human body picture, and the target vehicle picture. Kind or more.
- the retrieval module 51 is further configured to: obtain at least two retrieval conditions; retrieve a target feature matching any one of the at least two retrieval conditions from the database.
- the retrieval conditions include at least one of identity retrieval conditions, face retrieval conditions, human body retrieval conditions, and vehicle retrieval conditions.
- the target feature is pre-associated with identity information, and the identity information is any one of ID card information, name information, and file information.
- the retrieval module 51 is further configured to: use the sample feature of any one of the at least two retrieval conditions as the clustering center, cluster the target feature in the database, and set the preset cluster center
- the target features in the range are used as target features that match the retrieval conditions.
- this application also provides another target motion trajectory construction device.
- FIG. 6, is the structure of another embodiment of the target motion trajectory construction device provided by this application. Schematic.
- the target motion trajectory construction device 600 of this embodiment includes a processor 61, a memory 62, an input and output device 63 and a bus 64.
- the processor 61, the memory 62, and the input/output device 63 are respectively connected to the bus 64, the memory 62 stores a computer program, and the processor 61 is used to execute the computer program to implement the target motion trajectory construction method of the foregoing embodiment.
- the processor 61 may also be referred to as a CPU (Central Processing Unit, central processing unit).
- the processor 61 may be an integrated circuit chip with signal processing capability.
- the processor 61 may also be a general-purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component .
- the processor 61 can also be a GPU (Graphics Processing Unit, graphics processor), also known as a display core, a visual processor, and a display chip. It is a type of computer, workstation, game console, and some mobile devices (such as tablet computers, A microprocessor for image calculations on smartphones, etc.).
- the purpose of GPU is to convert and drive the display information required by the computer system, and to provide line scan signals to the display to control the correct display of the display. It is an important component for connecting the display to the PC motherboard and an important device for "human-machine dialogue". one.
- the graphics card is responsible for the task of outputting and displaying graphics. The graphics card is very important for those engaged in professional graphics design.
- the general-purpose processor may be a microprocessor or the processor 51 may also be any conventional processor or the like.
- the present application also provides a computer-readable storage medium.
- the computer-readable storage medium 700 is used to store a computer program 71.
- the computer program 71 is executed by a processor, it is used to achieve the target motion trajectory of this application Construct the method described in the method embodiment.
- the methods involved in the embodiments of the method for constructing the target motion trajectory of this application exist in the form of software functional units when implemented and when sold or used as independent products, they can be stored in the device, such as a computer readable storage medium.
- the technical solution of the present application essentially or the part that contributes to the existing technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , Including several instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor (processor) execute all or part of the steps of the methods described in the various embodiments of the present invention.
- the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical disks and other media that can store program codes. .
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Library & Information Science (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Image Analysis (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
目标/目标组合 | 检索方式 |
人脸 | 人脸检索、人脸人体融合检索 |
人体 | 人体融合检索 |
车辆 | 车辆检索 |
人脸+人体 | 人脸检索、人体融合检索 |
人脸+车辆 | 人脸检索、人脸融合检索、车辆检索 |
人体+车辆 | 人体融合检索、车辆检索 |
人脸+人体+车辆 | 人脸检索、人体融合检索、车辆检索 |
Claims (25)
- 一种目标运动轨迹构建方法,其中,所述方法包括:获取与检索条件匹配的至少两种不同类型的目标特征,其中所述至少两种不同类型的目标特征至少包括人脸特征、人体特征以及车辆特征中的至少两种;获取分别与所述至少两种不同类型的目标特征关联的拍摄时间和拍摄地点;根据所述至少两种不同类型的目标特征关联的拍摄时间和拍摄地点的组合生成目标运动轨迹。
- 根据权利要求1所述的方法,其中,所述根据所述至少两种不同类型的目标特征关联的拍摄时间和拍摄地点的组合生成目标运动轨迹的步骤进一步包括:以所述至少两种不同类型的目标特征中的某一类型的目标特征作为主目标特征,并以其他类型的目标特征为辅目标特征;根据所述主目标特征的拍摄时间和拍摄地点,以及所述辅目标特征的拍摄时间和拍摄地点判断所述辅目标特征与所述主目标特征相对位置是否符合目标的运动规律;若不符合目标的运动规律,剔除所述辅目标特征所关联的拍摄时间和拍摄地点。
- 根据权利要求2所述的方法,其中,所述根据所述主目标特征的拍摄时间和拍摄地点,以及所述辅目标特征的拍摄时间和拍摄地点判断所述辅目标特征与所述主目标特征相对位置是否符合目标的运动规律的步骤进一步包括:根据所述主目标特征的拍摄地点以及所述辅目标特征的拍摄地点计算位置差;根据所述主目标特征的拍摄时间以及所述辅目标特征的拍摄时间计算时间差;基于所述位置差和所述时间差计算运动速度,当所述运动速度大于预设运动速度阈值时,判断所述辅目标特征与所述主目标特征相对位置不符合目标的运动规律。
- 根据权利要求1或2所述的方法,其中,所述获取分别与所述至少两种不同类型的目标特征关联的拍摄时间和拍摄地点,包括:获取分别与所述至少两种不同类型的目标特征对应的第一目标图片;至少基于所述第一目标图片确定所述目标特征关联的拍摄时间和拍摄地点。
- 根据权利要求4所述的方法,其中,所述获取分别与所述至少两种不同类型的目标特征关联的第一目标图片之后,所述方法还包括:分别获取所述人脸特征对应的目标人脸图片,所述人体特征对应的目标人体图片和/或所述车辆特征对应的目标车辆图片;在所述目标人脸图片与所述目标人体图片对应于同一第一目标图片且具有预设空间关系 的情况下,将所述第一目标图片中的所述目标人脸图片与所述目标人体图片关联;在所述目标人脸图片与所述目标车辆图片对应于同一第一目标图片且具有预设空间关系的情况下,将所述第一目标图片中的所述目标人脸图片与所述目标车辆图片关联;在所述目标人体图片与所述目标车辆图片对应于同一第一目标图片且具有预设空间关系的情况下,将所述第一目标图片中的所述目标人体图片与所述目标车辆图片关联。
- 根据权利要求5所述的方法,其中,在所述至少两种不同类型的目标特征包括所述人脸特征的情况下,以及将所述第一目标图片中的所述目标人脸图片与所述目标车辆图片关联之后,所述方法还包括:基于所述目标车辆图片获取与所述目标车辆图片对应的第二目标图片;所述至少基于所述第一目标图片确定所述目标特征关联的拍摄时间和拍摄地点,包括:基于所述第一目标图片和所述第二目标图片确定所述目标特征关联的拍摄时间和拍摄地点。
- 根据权利要求5所述的方法,其中,在所述至少两种不同类型的目标特征包括所述人脸特征的情况下,以及将所述第一目标图片中的所述目标人脸图片与所述目标人体图片关联之后,所述方法还包括:基于所述目标人体图片获取与所述目标人体图片对应的第三目标图片;所述至少基于所述第一目标图片确定所述目标特征关联的拍摄时间和拍摄地点,包括:基于所述第一目标图片和所述第三目标图片确定所述目标特征关联的拍摄时间和拍摄地点。
- 根据权利要求5-7任一项所述的方法,其中,所述预设空间关系包括以下至少一种:所述第一目标关联图片的图像覆盖范围包含所述第二目标关联图片的图像覆盖范围;所述第一目标关联图片的图像覆盖范围与所述第二目标关联图片的图像覆盖范围部分重叠;所述第一目标关联图片的图像覆盖范围与所述第二目标关联图片的图像覆盖范围相连接;其中,所述第一目标关联图片包括所述目标人脸图片、所述目标人体图片及所述目标车辆图片中的任一一种或多种,所述第二目标关联图片包括所述目标人脸图片、所述目标人体图片及所述目标车辆图片中的任意一种或多种。
- 根据权利要求1所述的方法,其中,所述获取与检索条件匹配的至少两种不同类型的目标特征的步骤包括:获取至少两个所述检索条件;从数据库中检索所述至少两个检索条件中的任意一检索条件相匹配的目标特征。
- 根据权利要求9所述的方法,其中,所述检索条件包括身份检索条件、人脸检索条件、人体检索条件以及车辆检索条件中的至少一个;其中,所述目标特征预先关联有身份信息,所述身份信息为身份证信息、姓名信息和档案信息中的任意一种。
- 根据权利要求9所述的方法,其中,所述从数据库中检索所述至少两个检索条件中的任意一检索条件相匹配的目标特征的步骤包括:以至少两个检索条件中的任意一检索条件的样本特征作为聚类中心,对所述数据库中的目标特征进行聚类,将所述聚类中心的预设范围内的目标特征作为与所述检索条件相匹配的目标特征。
- 一种目标运动轨迹构建设备,其中,所述设备包括检索模块、获取模块以及轨迹构建模块;所述检索模块,用于获取与检索条件匹配的至少两种不同类型的目标特征,其中所述至少两种不同类型的目标特征至少包括人脸特征、人体特征以及车辆特征中的至少两种;所述获取模块,用于获取分别与所述至少两种不同类型的目标特征关联的拍摄时间和拍摄地点;所述轨迹构建模块,用于根据所述至少两种不同类型的目标特征关联的拍摄时间和拍摄地点的组合生成目标运动轨迹。
- 根据权利要求12所述的目标运动轨迹构建设备,其中,所述轨迹构建模块还用于:以所述至少两种不同类型的目标特征中的某一类型的目标特征作为主目标特征,并以其他类型的目标特征为辅目标特征;根据所述主目标特征的拍摄时间和拍摄地点,以及所述辅目标特征的拍摄时间和拍摄地点判断所述辅目标特征与所述主目标特征相对位置是否符合目标的运动规律;若不符合目标的运动规律,剔除所述辅目标特征所关联的拍摄时间和拍摄地点。
- 根据权利要求13所述的目标运动轨迹构建设备,其中,所述轨迹构建模块,还用于:根据所述主目标特征的拍摄地点以及所述辅目标特征的拍摄地点计算位置差;根据所述主目标特征的拍摄时间以及所述辅目标特征的拍摄时间计算时间差;基于所述位置差和所述时间差计算运动速度,当所述运动速度大于预设运动速度阈值时,判断所述辅目标特征与所述主目标特征相对位置不符合目标的运动规律。
- 根据权利要求12或13所述的目标运动轨迹构建设备,其中,所述获取模块还用于:获取分别与所述至少两种不同类型的目标特征对应的第一目标图片;至少基于所述第一目标图片确定所述目标特征关联的拍摄时间和拍摄地点。
- 根据权利要求15所述的目标运动轨迹构建设备,其中,所述获取模块还用于:分别获取所述人脸特征对应的目标人脸图片,所述人体特征对应的目标人体图片和/或所述车辆特征对应的目标车辆图片;在所述目标人脸图片与所述目标人体图片对应于同一第一目标图片且具有预设空间关系的情况下,将所述第一目标图片中的所述目标人脸图片与所述目标人体图片关联;在所述目标人脸图片与所述目标车辆图片对应于同一第一目标图片且具有预设空间关系的情况下,将所述第一目标图片中的所述目标人脸图片与所述目标车辆图片关联;在所述目标人体图片与所述目标车辆图片对应于同一第一目标图片且具有预设空间关系的情况下,将所述第一目标图片中的所述目标人体图片与所述目标车辆图片关联。
- 根据权利要求16所述的目标运动轨迹构建设备,其中,在所述至少两种不同类型的目标特征包括所述人脸特征的情况下,以及将所述第一目标图片中的所述目标人脸图片与所述目标车辆图片关联之后,所述获取模块还用于:基于所述目标车辆图片获取与所述目标车辆图片对应的第二目标图片;以及基于所述第一目标图片和所述第二目标图片确定所述目标特征关联的拍摄时间和拍摄地点。
- 根据权利要求16所述的目标运动轨迹构建设备,其中,在所述至少两种不同类型的目标特征包括所述人脸特征的情况下,以及将所述第一目标图片中的所述目标人脸图片与所述目标人体图片关联之后,所述获取模块还用于:基于所述目标人体图片获取与所述目标人体图片对应的第三目标图片;以及基于所述第一目标图片和所述第三目标图片确定所述目标特征关联的拍摄时间和拍摄地点。
- 根据权利要求16-18任一项所述的目标运动轨迹构建设备,其中,所述预设空间关系包括以下至少一种:所述第一目标关联图片的图像覆盖范围包含所述第二目标关联图片的图像覆盖范围;所述第一目标关联图片的图像覆盖范围与所述第二目标关联图片的图像覆盖范围部分重叠;所述第一目标关联图片的图像覆盖范围与所述第二目标关联图片的图像覆盖范围相连接;其中,所述第一目标关联图片包括所述目标人脸图片、所述目标人体图片及所述目标车 辆图片中的任一一种或多种,所述第二目标关联图片包括所述目标人脸图片、所述目标人体图片及所述目标车辆图片中的任意一种或多种。
- 根据权利要求12所述的目标运动轨迹构建设备,其中,所述检索模块还用于:获取至少两个所述检索条件;从数据库中检索所述至少两个检索条件中的任意一检索条件相匹配的目标特征。
- 根据权利要求20所述的目标运动轨迹构建设备,其中,所述检索条件包括身份检索条件、人脸检索条件、人体检索条件以及车辆检索条件中的至少一个;其中,所述目标特征预先关联有身份信息,所述身份信息为身份证信息、姓名信息和档案信息中的任意一种。
- 根据权利要求20所述的目标运动轨迹构建设备,其中,检索模块还用于:以至少两个检索条件中的任意一检索条件的样本特征作为聚类中心,对所述数据库中的目标特征进行聚类,将所述聚类中心的预设范围内的目标特征作为与所述检索条件相匹配的目标特征。
- 一种目标运动轨迹构建设备,其中,所述设备包括处理器和存储器;所述存储器中存储有计算机程序,所述处理器用于执行所述计算机程序以实现如权利要求1-11中任一项所述目标运动轨迹构建方法的步骤。
- 一种计算机可读存储介质,其中,所述计算机可读存储介质存储有计算机程序,所述计算机程序被执行时实现如权利要求1-11中任一项所述目标运动轨迹构建方法的步骤。
- 一种计算机程序产品,当所述计算机程序产品中的指令由处理器执行时,执行如权利要求1-11中任一所述的基于图片的目标运动轨迹构建方法。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020227020877A KR20220098030A (ko) | 2019-12-30 | 2020-07-03 | 타깃 운동 궤적 구축 방법, 기기 및 컴퓨터 저장 매체 |
JP2022535529A JP2023505864A (ja) | 2019-12-30 | 2020-07-03 | ターゲット移動軌跡の構築方法、機器及びコンピュータ記憶媒体 |
US17/836,288 US20220301317A1 (en) | 2019-12-30 | 2022-06-09 | Method and device for constructing object motion trajectory, and computer storage medium |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911402892.7A CN111400550A (zh) | 2019-12-30 | 2019-12-30 | 一种目标运动轨迹构建方法、设备以及计算机存储介质 |
CN201911402892.7 | 2019-12-30 |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US17/836,288 Continuation US20220301317A1 (en) | 2019-12-30 | 2022-06-09 | Method and device for constructing object motion trajectory, and computer storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2021135138A1 true WO2021135138A1 (zh) | 2021-07-08 |
Family
ID=71428378
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2020/100265 WO2021135138A1 (zh) | 2019-12-30 | 2020-07-03 | 一种目标运动轨迹构建方法、设备以及计算机存储介质 |
Country Status (6)
Country | Link |
---|---|
US (1) | US20220301317A1 (zh) |
JP (1) | JP2023505864A (zh) |
KR (1) | KR20220098030A (zh) |
CN (1) | CN111400550A (zh) |
TW (1) | TW202125332A (zh) |
WO (1) | WO2021135138A1 (zh) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112364722A (zh) * | 2020-10-23 | 2021-02-12 | 岭东核电有限公司 | 核电作业人员监控处理方法、装置和计算机设备 |
CN112883214B (zh) * | 2021-01-07 | 2022-10-28 | 浙江大华技术股份有限公司 | 特征检索方法、电子设备及存储介质 |
CN114543674B (zh) * | 2022-02-22 | 2023-02-07 | 成都睿畜电子科技有限公司 | 一种基于图像识别的检测方法及系统 |
CN114724122B (zh) * | 2022-03-29 | 2023-10-17 | 北京卓视智通科技有限责任公司 | 一种目标追踪方法、装置、电子设备及存储介质 |
CN114863400B (zh) * | 2022-04-06 | 2024-09-10 | 浙江大华技术股份有限公司 | 一种确定车辆轨迹的方法、装置、电子设备及存储介质 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140153908A1 (en) * | 2012-12-05 | 2014-06-05 | Canon Kabushiki Kaisha | Reproduction control apparatus, reproduction control method, and storage medium |
CN105975633A (zh) * | 2016-06-21 | 2016-09-28 | 北京小米移动软件有限公司 | 运动轨迹的获取方法及装置 |
CN109189972A (zh) * | 2018-07-16 | 2019-01-11 | 高新兴科技集团股份有限公司 | 一种目标行踪确定方法、装置、设备及计算机存储介质 |
CN110532923A (zh) * | 2019-08-21 | 2019-12-03 | 深圳供电局有限公司 | 一种人物轨迹检索方法及其系统 |
CN110532432A (zh) * | 2019-08-21 | 2019-12-03 | 深圳供电局有限公司 | 一种人物轨迹检索方法及其系统、计算机可读存储介质 |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9176987B1 (en) * | 2014-08-26 | 2015-11-03 | TCL Research America Inc. | Automatic face annotation method and system |
CN108875548B (zh) * | 2018-04-18 | 2022-02-01 | 科大讯飞股份有限公司 | 人物轨迹生成方法及装置、存储介质、电子设备 |
CN110070005A (zh) * | 2019-04-02 | 2019-07-30 | 腾讯科技(深圳)有限公司 | 图像目标识别方法、装置、存储介质及电子设备 |
CN110609916A (zh) * | 2019-09-25 | 2019-12-24 | 四川东方网力科技有限公司 | 视频图像数据检索方法、装置、设备和存储介质 |
-
2019
- 2019-12-30 CN CN201911402892.7A patent/CN111400550A/zh active Pending
-
2020
- 2020-07-03 WO PCT/CN2020/100265 patent/WO2021135138A1/zh active Application Filing
- 2020-07-03 KR KR1020227020877A patent/KR20220098030A/ko active Search and Examination
- 2020-07-03 JP JP2022535529A patent/JP2023505864A/ja not_active Withdrawn
- 2020-07-10 TW TW109123414A patent/TW202125332A/zh unknown
-
2022
- 2022-06-09 US US17/836,288 patent/US20220301317A1/en not_active Abandoned
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140153908A1 (en) * | 2012-12-05 | 2014-06-05 | Canon Kabushiki Kaisha | Reproduction control apparatus, reproduction control method, and storage medium |
CN105975633A (zh) * | 2016-06-21 | 2016-09-28 | 北京小米移动软件有限公司 | 运动轨迹的获取方法及装置 |
CN109189972A (zh) * | 2018-07-16 | 2019-01-11 | 高新兴科技集团股份有限公司 | 一种目标行踪确定方法、装置、设备及计算机存储介质 |
CN110532923A (zh) * | 2019-08-21 | 2019-12-03 | 深圳供电局有限公司 | 一种人物轨迹检索方法及其系统 |
CN110532432A (zh) * | 2019-08-21 | 2019-12-03 | 深圳供电局有限公司 | 一种人物轨迹检索方法及其系统、计算机可读存储介质 |
Also Published As
Publication number | Publication date |
---|---|
CN111400550A (zh) | 2020-07-10 |
US20220301317A1 (en) | 2022-09-22 |
TW202125332A (zh) | 2021-07-01 |
KR20220098030A (ko) | 2022-07-08 |
JP2023505864A (ja) | 2023-02-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2021135138A1 (zh) | 一种目标运动轨迹构建方法、设备以及计算机存储介质 | |
US9560323B2 (en) | Method and system for metadata extraction from master-slave cameras tracking system | |
US20220092881A1 (en) | Method and apparatus for behavior analysis, electronic apparatus, storage medium, and computer program | |
US11527000B2 (en) | System and method for re-identifying target object based on location information of CCTV and movement information of object | |
WO2021139324A1 (zh) | 图像识别方法、装置、计算机可读存储介质及电子设备 | |
US20210382933A1 (en) | Method and device for archive application, and storage medium | |
WO2021051545A1 (zh) | 基于行为识别模型的摔倒动作判定方法、装置、计算机设备及存储介质 | |
CN111488855A (zh) | 疲劳驾驶检测方法、装置、计算机设备和存储介质 | |
CN109902681B (zh) | 用户群体关系确定方法、装置、设备及存储介质 | |
US20210319226A1 (en) | Face clustering in video streams | |
KR20180015101A (ko) | 소스 비디오 내에서 관심 동영상을 추출하는 장치 및 방법 | |
CN109002776B (zh) | 人脸识别方法、系统、计算机设备和计算机可读存储介质 | |
CN111753766B (zh) | 一种图像处理方法、装置、设备及介质 | |
CN110619280B (zh) | 一种基于深度联合判别学习的车辆重识别方法及装置 | |
US20220027406A1 (en) | Method and system for using geographic information to direct video | |
JPWO2018179119A1 (ja) | 映像解析装置、映像解析方法およびプログラム | |
US20230008356A1 (en) | Video processing apparatus, method and computer program | |
CN112329665B (zh) | 一种人脸抓拍系统 | |
CN114913470A (zh) | 一种事件检测方法及装置 | |
Teja et al. | Man-on-man brutality identification on video data using Haar cascade algorithm | |
WO2021017289A1 (zh) | 在视频中定位对象的方法、装置、计算机设备及存储介质 | |
JP2022534314A (ja) | ピクチャに基づいた多次元情報の統合方法及び関連機器 | |
JP7540500B2 (ja) | グループ特定装置、グループ特定方法、及びプログラム | |
Golda | Image-based Anomaly Detection within Crowds | |
Kim et al. | SlowFast Based Real-Time Human Motion Recognition with Action Localization. |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20909155 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2022535529 Country of ref document: JP Kind code of ref document: A |
|
ENP | Entry into the national phase |
Ref document number: 20227020877 Country of ref document: KR Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 28/10/2022) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20909155 Country of ref document: EP Kind code of ref document: A1 |