An image compression method comprises: performing feature extraction on the target image to obtain a first feature map comprising a plurality of channels; grouping the channels of the first feature map to obtain a plurality of second feature maps; performing spatial context feature extraction on the second feature maps to determine first spatial redundancy features corresponding to the second feature maps; and performing channel context feature extraction on the second feature maps to determine first channel redundancy features corresponding to the second feature maps; determining compression information corresponding to each of the second feature maps based on a first spatial redundancy feature and a first channel redundancy feature corresponding to each of the second feature maps and thus determining first compressed data corresponding to the target image, and performing deep compression processing based on the first feature map to determine second compressed data corresponding to the target image.
H04N 19/136 - Incoming video signal characteristics or properties
H04N 19/42 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation
2.
METHOD FOR VIDEO PROCESSING, ELECTRONIC DEVICE, AND STORAGE MEDIUM
A video processing method, an electronic device and a storage medium are provided. The method includes: a target frame is determined in a video to be processed; at least two first energy images corresponding to the target frame are determined based on at least two preset macro block sizes respectively, each first energy image represents alternating current energy of first macro block(s) corresponding to a respective macro block size and the first macro block(s) are obtained by segmenting the target frame based on the respective macro block size; a first energy map corresponding to the target frame is determined based on the first energy images, the first energy map represents energy distribution in the target frame; and an adaptive quantization parameter corresponding to the target frame is determined based on the first energy map and the target frame is encoded using the adaptive quantization parameter.
H04N 19/119 - Adaptive subdivision aspects e.g. subdivision of a picture into rectangular or non-rectangular coding blocks
H04N 19/14 - Coding unit complexity, e.g. amount of activity or edge presence estimation
H04N 19/172 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a picture, frame or field
H04N 19/176 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
3.
POSE ESTIMATION METHOD AND APPARATUS, RELATED MODEL TRAINING METHOD AND APPARATUS, ELECTRONIC DEVICE, COMPUTER READABLE MEDIUM AND COMPUTER PROGRAM PRODUCT
Disclosed in the present disclosure are a pose estimation method and method, a related model training method and apparatus, an electronic device, a computer readable storage medium and a computer program product. The pose estimation model training method comprises: acquiring a sample image comprising an object to be positioned, the sample image comprising a sample color image and a sample depth image corresponding to the sample color image; processing, by using a pose estimation model, the sample color image, so as to obtain an initial sample pose of said object; optimizing the initial sample pose on the basis of depth information of said object in the sample depth image, so as to obtain an optimized pose of said object; and adjusting network parameters in the pose estimation model on the basis of the difference between the optimized pose and the initial sample pose.
A speaking state recognition method and apparatus, a model training method and apparatus, a vehicle, a medium, a computer program and a computer program product. The speaking state recognition method comprises: acquiring a facial image frame sequence of a target object (S101); acquiring mouth key point information of each image frame in the facial image frame sequence (S102); on the basis of the mouth key point information, determining a displacement feature of a mouth key point corresponding to the facial image frame sequence, the displacement feature representing a position change of the mouth key point between a plurality of image frames in the facial image frame sequence (S103); and determining a recognition result of the speaking state of the target object according to the displacement feature (S104).
Embodiments of the present disclosure provide a breathing state detection method and apparatus, a device, a storage medium and a computer program product. The method comprises: obtaining a chest image sequence, the chest image sequence comprising a plurality of frames of chest images of a target subject within a preset time period; obtaining a chest fluctuation signal according to the change of pixel values of the chest images in the chest image sequence within the preset time period; and determining a breathing state of the target subject according to the chest fluctuation signal.
Provided in the present disclosure are a traffic marking identification method and apparatus, a computer device and a storage medium. The method comprises: acquiring road point cloud data; performing grid partitioning on the road point cloud data, so as to obtain local point cloud data included in at least one target grid; and on the basis of the local point cloud data included in each target grid, determining a probability that the target grid belongs to a traffic marking category; and on the basis of the probability that target grids respectively belong to traffic marking categories, determining a marking identification result of the road point cloud data.
The present invention relates to the technical field of transmission. Provided in the embodiments of the present disclosure are a transmission wheel assembly and a mechanical arm. The transmission wheel assembly provided in the embodiments of the present disclosure comprises a base, a driving wheel, a driven wheel, and an end portion limiting member, wherein the driving wheel is rotatably connected to the base; the driven wheel is in transmission connection with the driving wheel by means of a flexible member, a first end of the driven wheel is rotatably connected to the base, a second end of the driven wheel is provided with a first projection, and the axis of the first projection coincides with the axis of rotation of the driven wheel; and the end portion limiting member is fixedly connected to the base and is provided with an abutting portion, the abutting portion is arranged close to an end portion of the first projection, so as to abut against the end portion of the first projection, thereby restricting movement of the driven wheel in the direction approaching the driving wheel, and the first projection is rotatable relative to the abutting portion. The transmission wheel assembly provided in the embodiments of the present disclosure is used for transmitting torque.
A tension mechanism, a transmission device and a mechanical arm. The tension mechanism (3) comprises a base (31), a tension wheel (32) and a pressure-applying assembly (33), wherein the tension wheel (32) is fixed relative to the base (31) and acts on a transmission portion of a transmission mechanism (2); and the pressure-applying assembly (33) is arranged on the base (31), and the pressure-applying assembly (33) comprises an input end (33211) and an output end (3311); the input end (33211) is in transmission connection with the output end (3311), and when the input end (33211) inputs a first pressure, the output end (3311) outputs a second pressure to the tension wheel (32), so as to provide a tension force for the tension wheel (32).
F16H 7/12 - Means for varying tension of belts, ropes, or chains by adjusting the axis of a pulley of an idle pulley
F16H 7/08 - Means for varying tension of belts, ropes, or chains
F16H 7/06 - Gearings for conveying rotary motion by endless flexible members with chains
F16H 7/02 - Gearings for conveying rotary motion by endless flexible members with beltsGearings for conveying rotary motion by endless flexible members with V-belts
9.
MECHANICAL ARM DRIVING STRUCTURE, ARM-TYPE ROBOT, DRIVING METHOD, STORAGE MEDIUM, AND COMPUTER PROGRAM PRODUCT
A mechanical arm driving structure, an arm-type robot, a driving method, a storage medium, and a computer program product. The mechanical arm driving structure comprises a driving member, a transmission mechanism, a monitoring unit, and a control unit; the transmission mechanism is transmittingly connected between the driving member and a mechanical arm, and is used for transmitting a driving force of the driving member to the mechanical arm and transmitting an external force received by the mechanical arm to the driving member; the monitoring unit is provided on the driving member, and is configured to acquire working parameters of the driving member, the working parameters changing along with the external force; the control unit is electrically connected to the driving member and the monitoring unit, and is configured to control a driving state of the driving member according to the working parameters; and the mechanical arm driving structure is used for driving the mechanical arm to move.
Embodiments of the present disclosure disclose an archive management method and apparatus, a device, a storage medium and a program product. The method comprises: on the basis of a plurality of frames of real-time images corresponding to a passenger in a vehicle, performing physiological state detection on the passenger; by means of a first interface of a connected display device, displaying real-time detection data obtained via the physiological state detection; when the physiological state detection meets a detection stopping condition, displaying the detection result of the passenger by means of a second interface of the display device, the detection result being determined on the basis of the real-time detection data; and in response to receiving a trigger operation for archive management control in the first interface or the second interface, by means of a third interface of the display device, managing archive data corresponding to the detection result.
G06F 16/11 - File system administration, e.g. details of archiving or snapshots
G06F 21/32 - User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
G16H 10/60 - ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
G16H 50/30 - ICT specially adapted for medical diagnosis, medical simulation or medical data miningICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indicesICT specially adapted for medical diagnosis, medical simulation or medical data miningICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for individual health risk assessment
11.
METHOD AND APPARATUS FOR INTERACTING WITH VEHICLE-MOUNTED DISPLAY DEVICE, AND DEVICE, STORAGE MEDIUM, AND COMPUTER PROGRAM PRODUCT
Disclosed in embodiments of the present invention are a method and apparatus for interacting with a vehicle-mounted display device, and the device, a storage medium, and a computer program product. The method comprises: when the vehicle-mounted display device works in a screensaver mode, obtaining a hand image of a person in a vehicle corresponding to the vehicle-mounted display device, wherein the vehicle-mounted display device displays, in the screensaver mode, a screensaver picture containing movable elements; determining, on the basis of first position information of a hand in the hand image, a target movable element matching the position of the hand among the movable elements of the screensaver picture; determining a dynamic interaction effect of the target movable element; and displaying the dynamic interaction effect of the target movable element in the screensaver picture.
Disclosed in embodiments of the present disclosure are a vehicle-mounted camera calibration method and apparatus, a computer device, a storage medium, and a product. The method comprises: obtaining first position information of a current position of a vehicle-mounted camera with respect to a reference position, wherein the vehicle-mounted camera is provided on a steering column; determining a first transformation matrix of the current position with respect to the reference position on the basis of the first position information of the vehicle-mounted camera; obtaining third position information of the reference position in a cabin coordinate system and rotation angle information of the steering column at the reference position; and determining a second transformation matrix of the current position of the vehicle-mounted camera with respect to the cabin coordinate system according to the first transformation matrix, the third position information, and the rotation angle information.
Disclosed are an eye type detection method and an apparatus, a computer device, a storage medium and a computer program product. The method comprises: in a case of determining that a target object faces an image acquisition device, acquiring a front face image of the target object acquired by the image acquisition device; determining a pupil position of the left eye and a pupil position of the right eye of the target object in the front face image, respectively; determining, according to the pupil position of the left eye, a gaze angle of the left eye, and determining, according to the pupil position of the right eye, a gaze angle of the right eye of the target object; determining an angle difference between the gaze angle of the left eye and the gaze angle of the right eye; in a case that the angle difference exceeds a first predetermined angle range, determining a larger one of absolute values of the gaze angles of the left eye and the right eye as a strabismus eye in the front face image; and according to the strabismus eye in the front face image, calibrating an eye strabismus type of the target object. According to the embodiment of the present disclosure, the eye strabismus type of the target object can be conveniently and accurately detected in an image processing mode.
A61B 3/08 - Subjective types, i.e. testing apparatus requiring the active assistance of the patient for testing binocular or stereoscopic vision, e.g. strabismus
14.
COMPILER TEST METHOD AND APPARATUS, CASE GENERATION METHOD AND APPARATUS, AND INSTRUCTION STORAGE STRUCTURE
A compiler test method and apparatus, a device, a storage medium, and a storage structure for a test instruction. The method comprises: on the basis of data which is stored in an instruction database and is related to a test instruction, constructing a test case, and a reference compiling instruction corresponding to the test case; compiling the test case by using a compiler to be tested, so as to obtain a target compiling instruction; and on the basis of a comparison result between the reference compiling instruction corresponding to the test case and the target compiling instruction, determining whether the compiler is abnormal. In the embodiments of the present disclosure, automatic testing of a compiler can be realized, thereby improving the test efficiency and the accuracy of a test result.
Disclosed in embodiments of the present disclosure are a physiological state detection method and apparatus, a device, a storage medium and a program product. The method comprises: in response to receiving a request for physiological state detection of a passenger in a vehicle, obtaining an image of the passenger acquired by means of a vehicle-mounted camera; detecting a physiological state of the passenger on the basis of the image to obtain the physiological state detection result; and analyzing the physiological state detection result, and when an abnormal index exists in the physiological state detection result, pushing an online health consultation service to the passenger.
G16H 50/30 - ICT specially adapted for medical diagnosis, medical simulation or medical data miningICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indicesICT specially adapted for medical diagnosis, medical simulation or medical data miningICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for individual health risk assessment
16.
IMAGE PROCESSING METHOD AND APPARATUS, MODEL GENERATION METHOD AND APPARATUS, VEHICLE, STORAGE MEDIUM, AND COMPUTER PROGRAM PRODUCT
Disclosed in embodiments of the present disclosure are an image processing method and apparatus, a model generation method and apparatus, a vehicle, a storage medium, and a computer program product. The image processing method comprises: obtaining an image frame sequence comprising a mouth object; extracting mouth key point features from each image frame in the image frame sequence to obtain the mouth key point features of each image frame; generating syllable classification features according to the mouth key point features of the plurality of image frames in the image frame sequence, wherein the syllable classification features each represent a syllable category corresponding to a mouth shape of the mouth object in the image frame sequence; and determining, in a preset keyword library, a keyword matched with the syllable classification features.
Embodiments of the present disclosure provide a data processing method and apparatus, a chip, an electronic device, and a medium. The data processing apparatus comprises a node array, and the node array comprises a plurality of node groups. For each of the plurality of node groups, adjacent nodes in the node group are connected, and each node in the node group is connected to a plurality of nodes of the other node groups. The plurality of nodes comprise default nodes and standby nodes, and a default node of a node is a standby node of at least one of the other nodes in the node group where the node is located. In the case that a default node of a first node in a node group is in a normal state, a connection between the first node and the corresponding default node is enabled; and in the case that a default node of a first node in a node group is in an abnormal state, connections between the first node and the corresponding standby node and between at least one second node in the same node group as the first node and the corresponding standby node are enabled.
G06F 11/20 - Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
The present disclosure relates to the technical field of chip verification, and particularly provides a chip verification method and platform. The chip verification platform comprises: an excitation data component, configured to generate excitation data on the basis of a case under test and send the excitation data to a component to be verified; and a counting control component, configured to detect and count, in real time, in a verification process, a target signal of each piece of result data output by the component to be verified, and determine, in response to the fact that the number of the target signals meets a preset condition, that verification of the case under test is finished, the target signal representing a timing dependent signal of the result data. According to embodiments of the present disclosure, the chip verification efficiency is improved.
G06F 30/398 - Design verification or optimisation, e.g. using design rule check [DRC], layout versus schematics [LVS] or finite element methods [FEM]
19.
MODEL TRAINING METHOD AND APPARATUS, METHOD AND APPARATUS FOR PREDICTING MODAL INFORMATION, AND ELECTRONIC DEVICE, STORAGE MEDIUM AND COMPUTER PROGRAM PRODUCT
The present disclosure relates to a model training method and apparatus, a method and apparatus for predicting modal information, and an electronic device, a storage medium and a computer program product. The model training method comprises: acquiring a prediction model, which is obtained by means of initial training, wherein the prediction model which is obtained by means of initial training is used for predicting information of an original mode; and retraining the prediction model by using preset sample data, wherein the preset sample data comprises first sample data, which is annotated with information of a target mode, and the retrained prediction model is further used for predicting the information of the target mode; and the preset sample data further comprises second sample data, which is annotated with the information of the original mode, and/or retraining the prediction model comprises: constraining the adjustment of at least some network parameters of the prediction model.
An action detection method and apparatus, an electronic device, and a storage medium. The action detection method comprises: acquiring a video stream of a scene area (S101); on the basis of a plurality of image frames in the video stream, detecting an action direction of a target object in the scene area (S102); detecting skeleton key points of the target object in the plurality of image frames (S103); and according to a geometric relationship between target key points among the detected skeleton key points corresponding to the action direction, determining action information of the target object (S104).
Provided are a data labeling method and apparatus, an electronic device, and a storage medium. The method comprises: obtaining a data set to be labeled, the data set having a pre-labeling result, and the pre-labeling result comprising a pre-labeling tag; in response to receiving a setting instruction for a training scene, according to the training scene indicated by the setting instruction, converting the pre-labeling tag into at least one level of attribute tag suitable for the training scene; and labeling data in the data set according to the at least one level of attribute tag to obtain a labeled target data set.
A data processing method and apparatus, an electronic device, and a storage medium. The data processing method comprises: acquiring an image data set to be processed (S11); in response to receiving a setting instruction for a data processing flow of the image data set, determining, according to data processing modes indicated by the setting instruction and an execution sequence of the data processing modes, the data processing flow corresponding to the image data set (S12); and processing image data in the image data set according to the execution sequence of the data processing modes in the data processing flow, to obtain a processed image data set (S13).
An application definition method and apparatus (90), an electronic device (10), a storage medium, and a computer program product. The method is applied to an intelligent front-end device, and comprises: according to hardware resource parameters of the intelligent front-end device, determining a target hardware resource occupation level matched with the intelligent front-end device; loading a program package of a first application under the target hardware resource occupation level to the intelligent front-end device according to a preset service interface framework in the intelligent front-end device, wherein the service interface framework at least comprises an interface used for running an application; and under the condition that the first application is run on the basis of the interface, which is used for running the application, in the service interface framework, the intelligent front-end device achieving a function corresponding to the first application.
Provided in the embodiments of the present disclosure are a processor, a chip, an electronic device, and a data processing method. The processor comprises: a first register file, wherein the first register file comprises at least one first register group and at least one second register group, each first register group and each second register group comprise at least one register, each first register group is allocated to one thread among a plurality of threads, and each second register group is allocated to at least two threads among the plurality of threads; and a processing unit, which is used for scheduling each thread among the plurality of threads, and accessing, in response to a data access request of a target thread among the plurality of threads, a target register in a register group which is allocated to the target thread, wherein the register group comprises the first register group or the second register group. The embodiments of the present disclosure implement data multiplexing between threads.
Provided in the embodiments of the present disclosure are a wear leveling method and apparatus for a memory, and a memory and an electronic device. The memory comprises: a plurality of storage units, which are used for storing data; and a first control unit, which is used for receiving a write-in request, wherein the write-in request carries an input address; the input address is mapped to a target address on the basis of an address mapping function of the input address, so as to write data into a storage unit corresponding to the target address; and an address mapping function of an input address is generated on the basis of an accumulative distribution feature of the number of times the input address is written, and the accumulative distribution feature is used for representing the accumulative number of times the input address and each previous input address prior to the input address in the memory are written.
Disclosed in the embodiments of the present disclosure are a model control method and apparatus, and a device, a storage medium and a computer program product. The method comprises: acquiring an image frame set to be processed, wherein there is a sequence timing relationship among image frames in the image frame set, and the image frames each comprise at least one first target object, which participates in a task; determining a matching relationship between the at least one first target object and at least one controlled model for executing the task; recognizing at least two image frames in said image frame set, so as to obtain a first recognition result, wherein the first recognition result at least comprises a first motion parameter of each first target object; and on the basis of the matching relationship, controlling a controlled model, which matches each first target object, to complete the task according to a second motion parameter, which matches the first motion parameter.
Embodiments of the present invention provide a physiological state detection method and apparatus, an electronic device, a storage medium, and a program. The method comprises: acquiring a video stream collected by a photographing device; extracting multiple frames of face images of a target object from multiple frames of images in the video stream; determining at least one smooth area in the face images; performing weighting processing or area screening on the at least one smooth area on the basis of area attribute information, so as to obtain information to be extracted that is related to a physiological state; and performing physiological state information extraction on the information to be extracted, so as to obtain a physiological state detection result of the target object. According to the embodiments of the present invention, the physiological state detection is implemented on the basis of an imaging processing mode, real-time measurement can be performed anytime and anywhere, the practicability is better, and the accuracy of measurement can be improved by means of the constraint of preset conditions.
Embodiments of the present invention provide a detection result processing method and apparatus, a device, a medium, and a computer program product. The detection result processing method comprises: obtaining a detection result of an image to be detected; determining, on the basis of a historical detection result, whether the detection result satisfies a preset condition; and in response to the detection result satisfying the preset condition, executing preset processing corresponding to the detection result.
Provided in the embodiments of the present disclosure are an interrupt controller, a chip, a computer device, an interrupt control method, and a medium. The interrupt controller comprises a configuration unit, a first selection unit and a normalization processing unit. The normalization processing unit comprises a plurality of normalization sub-units, wherein different normalization sub-units use different normalization processing modes. The configuration unit is used for acquiring first configuration information, wherein the first configuration information is pre-generated on the basis of the type of an interrupt request, and each type of interrupt request corresponds to a normalization processing mode. The first selection unit is used for sending the interrupt request to one normalization sub-unit among the plurality of normalization sub-units on the basis of the first configuration information, such that the selected normalization sub-unit performs normalization processing on the received interrupt request in the normalization processing mode corresponding to this unit.
Provided in the embodiments of the present disclosure are a method and apparatus for tracking the state of an article, and an electronic device, a storage medium and a computer program product. The method comprises: acquiring movement trajectory information of an article, wherein the movement trajectory information includes detection results of the article in at least two frames of images; on the basis of the movement trajectory information and information regarding a preset detection line, determining detection state information of the article in each frame of image, wherein the detection line is arranged on one side of a storage cabinet, the distance between the detection line and the storage cabinet is within a preset range, and the detection state information is used for measuring the position change state of the article relative to the detection line; and on the basis of detection state information of the article in every two adjacent frames of images from among the at least two frames of images, determining a target tracking state corresponding to the article, wherein the target tracking state is used for indicating that the article has been taken or that the article has been returned.
G06V 20/40 - ScenesScene-specific elements in video content
G06T 7/254 - Analysis of motion involving subtraction of images
G07F 17/12 - Coin-freed apparatus for hiring articlesCoin-freed facilities or services for means for safe-keeping of property, left temporarily, e.g. by fastening the property comprising lockable containers, e.g. for accepting clothes to be cleaned
31.
TARGET RECOGNITION, AND TRAINING OF NEURAL NETWORK
The present disclosure relates to target recognition, and training of a neural network for target recognition. According to an example of the present disclosure, a method comprises: a second client in a neural network training system, which at least comprises a first client and the second client, acquiring a first neural network obtained by the first client by means of using first sample image data, and then, inputting second sample image data into the first neural network, so as to obtain first feature data, which is output by the first neural network; according to the second sample image data and the first feature data, training a second neural network to be trained, so as to obtain a trained second neural network; and training a target neural network, which comprises the trained second neural network and the first neural network, by using the second sample image data, until a convergence condition is met.
A medical data management method and system, a device, a medium, and a computer program product. The medical data management method comprises: detecting a login to a data management platform, wherein the data management platform is used for managing medical data of a target object, and the medical data of the target object comprises at least one of clinical data of the target object and bioomics data of the target object; and in response to a preset operation on the data management platform, performing data processing matching the preset operation on the medical data. According to the medical data management method, a user can conveniently manage medical data in a data management platform.
The present disclosure relates to an action detection method and apparatus, an electronic device and a storage medium. The action detection method comprises: obtaining an image of a target object; performing skeleton key point detection on the image of the target object to obtain skeleton key point detection information of the target object; performing face key point detection on the image of the target object to obtain face key point detection information of the target object; obtaining an estimation result of action information of the target object according to a geometrical relationship between target key points in the skeleton key point detection information; and verifying the estimation result according to the face key point detection information to obtain an action detection result of the target object.
The present disclosure relates to a three-dimensional data display method and apparatus, an electronic device and a storage medium. The method comprises: obtaining three-dimensional scanning data of a human body part, the three-dimensional scanning data comprising slice data of the human body part in three dimensions and slice position information of the slice data of the three dimensions corresponding to the human body part; determining the slice data at a target slice position according to the target slice positions specified in the three dimensions respectively and the slice position information; and displaying a three-dimensional slice model of the human body part according to the slice data at the target slice position, the three-dimensional slice model comprising a slice map of the human body part at the target slice position obtained by rendering according to the slice data.
A liveness detection method and apparatus, and an electronic device, a storage medium, a computer program and a computer program product. The method comprises: acquiring an infrared facial image and a color facial image, that are collected by a binocular camera, of an object subjected to detection (201); performing feature extraction on the infrared facial image so as to obtain a first facial feature map, and performing feature extraction on the color facial image so as to obtain a second facial feature map (202); obtaining target category attribute information of said object on the basis of the second facial feature map (203); obtaining a third facial feature map on the basis of the target category attribute information and the second facial feature map (204); and obtaining a liveness detection result of said object on the basis of the first facial feature map and the third facial feature map (205).
Disclosed in the embodiments of the present disclosure are an interaction method and apparatus, and a device, a storage medium and a computer program product. The method comprises: acquiring a scene image; performing person recognition on the scene image, so as to obtain a recognition result, wherein the recognition result comprises person information of each person object in the scene image; when the recognition result represents that at least two person objects appear in the scene image, on the basis of the person information of each person object, determining an interaction message corresponding to at least one target person object; and displaying, in the scene image, the interaction message corresponding to the at least one target person object.
A chess placement method and apparatus, an electronic device, a chess playing robot, a computer storage medium, and a computer program. The chess placement method comprises: controlling an image acquisition device to acquire a first image, the first image representing an image of a chessboard and each chess piece; in response to the existence of a blocking area in a chessboard image area in the first image, determining a first area in the area other than the blocking area in the chessboard image area; in response a chess piece corresponding to at least one correct placement position is not placed in the at least one correct placement position in the first area, generating a chess placement instruction according to the at least one correct placement position and the actual position of at least one chess piece in the first image, the correct placement position representing a chess piece placement position determined according to a chess piece placement rule; and controlling an executor of a chess playing robot to execute the chess placement instruction.
An image analysis method and apparatus, a model training method and apparatus, and a device, a medium and a program. The image analysis method comprises: acquiring an image sequence, optical flow data, and reference data of each image in the image sequence (S11), wherein each image comprises a first image and a second image, which have a co-visibility relationship, the optical flow data comprises a static optical flow and an overall optical flow between the first image and the second image, the static optical flow is caused by the movement of a photographic device, the overall optical flow is caused by both the movement of the photographic device and the movement of a photographic subject, and the reference data comprises a pose and a depth; on the basis of the image sequence and the optical flow data, performing prediction to obtain an analysis result (S12), wherein the analysis result comprises optical flow calibration data of the static optical flow; and on the basis of the static optical flow and the optical flow calibration data, optimizing the pose and the depth, so as to obtain an updated pose and an updated depth (S13). By means of the solution, the precision of a pose and a depth can be improved in a dynamic scenario.
Provided in the present disclosure are an interaction control method and apparatus, and a computer device and a storage medium. The method comprises: in response to a target interaction request being acquired, acquiring a first color sequence, wherein the first color sequence comprises a plurality of colors; on the basis of the first color sequence, acquiring first facial images, which respectively correspond to a target face in the plurality of colors in the first color sequence; on the basis of the first color sequence, performing color detection and living body detection on the first facial images, and on the basis of the first facial images, performing identity detection; and in response to the first facial images passing the color detection, the living body detection and the identity detection, executing an interaction operation corresponding to the target interaction request.
The present disclosure relates to an image fusion method and apparatus, and an electronic device, a storage medium, a computer program and a computer program product. The method comprises: acquiring a first image and a second image, which are to be fused, wherein there is the same type of objects in the first image and the second image; respectively performing encoding processing on the first image and the second image, so as to obtain a first hidden variable corresponding to the first image and a second hidden variable corresponding to the second image; in response to a setting operation for a fusion weight of any object attribute of the same type of objects, fusing the first hidden variable with the second hidden variable according to the set fusion weight, so as to obtain a fused third hidden variable; and performing decoding processing on the third hidden variable, so as to obtain a fused target image. By means of the embodiments of the present disclosure, a fused target image can meet different fusion requirements of users.
The present disclosure relates to an image processing method and apparatus, and an electronic device, a storage medium and a program product. The method comprises: acquiring a facial image to be processed; performing encoding processing on the facial image, so as to obtain a first hidden variable of the facial image; in response to a setting operation for an attribute editing degree of a facial attribute, editing the first hidden variable according to the set attribute editing degree and an attribute editing direction corresponding to the facial attribute, so as to obtain an edited second hidden variable, wherein the attribute editing direction represents an enhancement direction or a weakening direction of the facial attribute, different facial attributes correspond to different attribute editing directions, and the attribute editing degree represents an enhancement degree or a weakening degree of the facial attribute; and performing decoding processing on the second hidden variable, so as to obtain a target facial image, wherein the display effect of the facial attribute in the target facial image is different from the display effect of the facial attribute in the facial image.
G06V 40/16 - Human faces, e.g. facial parts, sketches or expressions
G06T 3/00 - Geometric image transformations in the plane of the image
G06V 10/764 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
42.
CROWD COUNTING METHOD AND APPARATUS, ELECTRONIC DEVICE AND STORAGE MEDIUM
The present disclosure relates to a crowd counting method and apparatus, an electronic device and a storage medium. The method comprises: obtaining a plurality of crowd images, wherein each crowd image corresponds to one image acquisition device, and each image acquisition device corresponds to one preset counting region; determining, on the basis of head key point positioning respectively performed on the plurality of crowd images, the number of people in the preset counting region corresponding to each image acquisition device; and determining the number of people in a target region according to the number of people in the preset counting region corresponding to each image acquisition device, wherein the target region represents a space range requiring crowd counting, and the target region comprises the preset counting region corresponding to each image acquisition device. According to embodiments of the present disclosure, the accuracy of crowd counting in a scenario with a large area and a dense crowd can be effectively improved.
Provided in the embodiments of the present disclosure are a programming method and apparatus, and a device, a storage medium and a computer program product. The programming method comprises: acquiring at least one code module related to a target project, wherein each code module respectively corresponds to at least one code segment; displaying a selection interface of the at least one code module; and in response to an instruction of selection made by a user for a code module by means of the selection interface, displaying a code segment of the selected code module, wherein the displayed code segment can be edited.
Embodiments of the present disclosure relate to a user recognition method and apparatus, a storage medium, an electronic device, a computer program product and a computer program. The method comprises: when obtaining first device information corresponding to at least one target terminal, obtaining a target information set according to at least one piece of first biological information associated with each piece of the first device information, wherein the target information set further comprises user information corresponding to each piece of the first biological information; when obtaining second biological information, querying in the target information set target biological information matched with the second biological information; and when obtaining the target biological information, determining a target user according to the user information corresponding to the target biological information.
Provided in the present disclosure are a facial recognition method and apparatus, and an electronic device, a storage medium, a computer program and a computer program product. The method comprises: acquiring a target signal, wherein the target signal is a communication signal, the duration of stay in a target area of which is greater than or equal to a preset duration; on the basis of a device identifier of a target device, acquiring preset facial information corresponding to the target signal, and caching the preset facial information in a local cache, wherein the target device is a device which sends the target signal; and comparing facial information to be recognized of an object to be recognized with the preset facial information in the local cache, so as to obtain a facial recognition result corresponding to the object to be recognized.
G06V 40/16 - Human faces, e.g. facial parts, sketches or expressions
G06F 16/583 - Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
G06Q 20/40 - Authorisation, e.g. identification of payer or payee, verification of customer or shop credentialsReview and approval of payers, e.g. check of credit lines or negative lists
46.
FACE RECOGNITION METHOD AND APPARATUS, ELECTRONIC DEVICE, STORAGE MEDIUM, COMPUTER PROGRAM, AND PROGRAM PRODUCT
The present disclosure relates to a face recognition method and apparatus, an electronic device, a storage medium, a computer program, and a program product. The method comprises: acquiring, at a face recognition time node, face information to be recognized of an object to be recognized; and comparing the face information to be recognized with face information in a locally stored target face information sub-library to obtain a face recognition result of the object to be recognized, wherein the target face information sub-library is a face information sub-library associated with the face recognition time node, and the amount of face information comprised in the target face information sub-library is smaller than the amount of face information comprised in a face information full library. The present disclosure can reduce the occupation of network resources during a face recognition process and improve the efficiency of face recognition.
An automatic account checking method and apparatus, and an electronic device and a storage medium. The method comprises: determining at least one account, and a billing rule and service information for each account (S10); generating a corresponding user bill according to the billing rule and the service information for each account (S20); respectively sending the corresponding user bill to each account (S30); determining whether each user bill has been confirmed (S40); and if each user bill has been confirmed, generating a total bill comprising all user bills, and sending the total bill to a financial system (S50). By using the method, a bill is automatically generated, and account checking is automatically performed according to the generated bill, thereby improving the efficiency and security of an account checking process.
G06Q 20/10 - Payment architectures specially adapted for electronic funds transfer [EFT] systemsPayment architectures specially adapted for home banking systems
48.
ATTENDANCE CHECK METHOD AND APPARATUS, AND ELECTRONIC DEVICE, STORAGE MEDIUM AND PROGRAM PRODUCT
An attendance check method and apparatus, and an electronic device, a storage medium and a program product. The method comprises: in response to detecting a gesture corresponding to any preset attendance check state, determining the preset attendance check state as a current attendance check state (S11); performing facial recognition on a facial image of the current person, so as to obtain identity information of the current person (S12); and performing attendance check processing according to the current attendance check state and the identity information (S13).
G07C 1/10 - Registering, indicating, or recording the time of events or elapsed time, e.g. time-recorders for work people together with the recording, indicating or registering of other data, e.g. of signs of identity
49.
FACE MATCHING METHOD AND APPARATUS, ELECTRONIC DEVICE, STORAGE MEDIUM, COMPUTER PROGRAM PRODUCT, AND COMPUTER PROGRAM
The present disclosure relates to a face matching method and apparatus, an electronic device, a storage medium, a computer program product, and a computer program. The method comprises: acquiring a face image to be matched; determining a first database and a second database, wherein the second database is a subset of the first database; if the amount of second face information comprised in the second database is greater than or equal to an amount threshold, and the popularity value of each piece of second face information is greater than or equal to a popularity threshold, performing face matching on the face image to be matched according to the second database, wherein the popularity value is used for representing the probability that the second face information is matched; and if the face matching fails, performing face matching on the face image to be matched according to the first database.
G06V 10/74 - Image or video pattern matchingProximity measures in feature spaces
G06V 40/16 - Human faces, e.g. facial parts, sketches or expressions
G06K 9/62 - Methods or arrangements for recognition using electronic means
G06F 16/583 - Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
50.
FACIAL RECOGNITION PAYMENT METHOD AND APPARATUS, ELECTRONIC DEVICE, STORAGE MEDIUM, PROGRAM AND PRODUCT
The present disclosure relates to a facial recognition payment method and apparatus, an electronic device, a storage medium, a program and a product. The method comprises: receiving a facial recognition payment request sent by a target payment terminal, wherein the facial recognition payment request comprising a face image to be verified; determining a target terminal group to which the target payment terminal belongs; performing identity verification on said face image on the basis of a target face image set corresponding to the target terminal group, to obtain a first identity verification result, wherein the target face image set is a registered face image set of a target object group that provides a payment service for the target terminal group; and under the condition that the first identity verification result indicates that the target face image set comprises a target face registration image matching the face image to be verified, executing a payment operation on the basis of a payment account corresponding to the target face registration image.
G06Q 20/40 - Authorisation, e.g. identification of payer or payee, verification of customer or shop credentialsReview and approval of payers, e.g. check of credit lines or negative lists
51.
DETECTION METHOD AND APPARATUS, ELECTRONIC DEVICE, AND STORAGE MEDIUM
The present disclosure relates to a detection method and apparatus, an electronic device and a storage medium. The detection method comprises: receiving an image sequence sent by a terminal in response to an action sequence, the image sequence comprising multiple frames of images; sequentially acquiring one action content in the action sequence as current action content, and performing the following operations: determining a starting image corresponding to the current action content, and sequentially determining action scores of each image after the starting image in the image sequence and of the current action content; determining, according to the action score of any image, a matching result corresponding to the current action content; determining a matching result between the image sequence and the action sequence according to the matching results corresponding to all the action content in the action sequence; and generating a detection result on the basis of the matching result between the image sequence and the action sequence. The present disclosure can improve the security of a user detection environment and the accuracy of detection results.
G06V 10/75 - Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video featuresCoarse-fine approaches, e.g. multi-scale approachesImage or video pattern matchingProximity measures in feature spaces using context analysisSelection of dictionaries
G06V 10/82 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
52.
STRUCTURED TARGET DETECTION METHOD AND APPARATUS, DEVICE, STORAGE MEDIUM, AND PROGRAM PRODUCT
Disclosed in embodiments of the present invention are a structured target detection method and apparatus, a device, a storage medium, and a program product. The method comprises: acquiring a batch of video frame images to be processed; performing target detection and tracking on at least two targets in the batch of video frame images to obtain a detection and tracking result of each of the targets; determining the association relationship between the at least two targets on the basis of the detection and tracking results of the at least two targets; on the basis of the detection and tracking result of each of the targets, determining, according to a preset frame selection policy, target snapshot images corresponding to the target, the frame selection policy being used for constraining the number of target snapshot images of the same target; and determining a structured target of the batch of video frame images on the basis of the association relationship between the at least two targets and the target snapshot images of each of the targets.
Embodiments of the present disclosure relate to a pose determination method and apparatus, an electronic device, a storage medium, and a program. The method comprises: determining at least one first feature point in an image to be recognized on a target object; determining at least one target image from among pre-stored images according to the target object in the image, wherein the target image has a second feature point and a corresponding three-dimensional point cloud, and the second feature point has a corresponding point in the three-dimensional point cloud; determining a target point of the at least one first feature point in the three-dimensional point cloud according to the at least one first feature point, the second feature point, and the point corresponding to the second feature point in the three-dimensional point cloud; and according to the target point, determining a target pose corresponding to the image. In the embodiments of the present disclosure, a three-dimensional feature point corresponding to each two-dimensional feature point in an image to be recognized can be determined by means of local feature matching, so that by means of the pose of an image collection apparatus when the image is collected according to the three-dimensional feature point, the accuracy of a determined target pose can be improved.
G06V 10/44 - Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersectionsConnectivity analysis, e.g. of connected components
G06V 10/75 - Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video featuresCoarse-fine approaches, e.g. multi-scale approachesImage or video pattern matchingProximity measures in feature spaces using context analysisSelection of dictionaries
G06V 10/25 - Determination of region of interest [ROI] or a volume of interest [VOI]
54.
IMAGE PREVIEW METHOD AND APPARATUS, ELECTRONIC DEVICE, STORAGE MEDIUM, COMPUTER PROGRAM, AND PRODUCT THEREOF
Embodiments of the present disclosure provide an image preview method and apparatus, an electronic device, a storage medium, a computer program, and a product thereof. The method comprises: obtaining target encoded data, the target encoded data being obtained after a target encoding network performs image encoding on a target image; and performing image decoding on the target encoded data by using a target preview network corresponding to the target encoding network so as to obtain a target preview image corresponding to the target image, wherein the resolution of the target preview image is less than the resolution of the target image.
TRAINING METHOD FOR VISUAL INERTIAL ODOMETER MODEL, POSTURE ESTIMATION METHOD AND APPARATUSES, ELECTRONIC DEVICE, COMPUTER-READABLE STORAGE MEDIUM, AND PROGRAM PRODUCT
The present disclosure discloses a training method for a visual inertial odometer model, a posture estimation method and apparatuses, an electronic device, a computer-readable storage medium, and a program product. The training method for a visual inertial odometer model comprises: inputting two adjacent frames of sample color images in a sample image set and sample IMU data corresponding to the two adjacent frames of sample color images into a visual inertial odometer model, and outputting two frames of depth images corresponding to the two adjacent frames of sample color images and an estimated posture when an image collection apparatus obtains the two adjacent frames of sample color images; determining an objective loss function of the visual inertial odometer model on the basis of the two frames of depth images corresponding to the two adjacent frames of sample color images, the estimated posture when the image collection apparatus obtains the two adjacent frames of sample color images, and the corresponding sample IMU data between the two adjacent frames of sample color images; and adjusting network parameters of the visual inertial odometer model by using the objective loss function. According to the foregoing solution, a robust posture estimation result can be obtained.
Disclosed in the embodiments of the present disclosure are a model quantization method and apparatus, a device, a storage medium and a computer program product, the method comprising: acquiring first output data of at least one first network substructure in a first network model after the first network model processes a calibration data set; by using a second network model obtained by quantizing the first network model, processing the calibration data set on the basis of an activation quantization identifier of at least one second network substructure in the second network model, so as to obtain second output data of each second network substructure; and for each first network substructure, adjusting, on the basis of the first output data of the first network substructure and the second output data of the second network substructure corresponding to the first network substructure, a parameter of the second network substructure; and when it is determined that a preset condition is met, determining the adjusted second network model as a third network model.
G06F 30/27 - Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
G06K 9/62 - Methods or arrangements for recognition using electronic means
G06F 119/02 - Reliability analysis or reliability optimisationFailure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
57.
IMAGE ANALYSIS METHOD AND APPARATUS, DEVICE, STORAGE MEDIUM, AND PROGRAM PRODUCT
The present application discloses an image analysis method and apparatus, a device, a storage medium, and a program product. The image analysis method comprises: obtaining first position information about a target in a medical image; determining candidate paths respectively from a plurality of target points to the target on the basis of the first position information, wherein each candidate path comprises an action range for the target; and selecting at least one final path from the plurality of candidate paths, wherein the sum of the action ranges comprised in the at least one final path can cover a preset range of the target. According to the solution, the efficiency of path planning can be improved.
Provided in the embodiments of the present disclosure are an image inpainting method and apparatus, and a device and a medium. The method comprises: acquiring a facial image of a target object and an environment image, which comprises the surrounding environment of the target object, wherein the facial image includes a spectacle lens area image of glasses worn by the target object; according to a matching result of the spectacle lens area image and the environment image, determining a light reflection area in the spectacle lens area image; and according to the light reflection area, inpainting the spectacle lens area image, so as to obtain an inpainted target image. By means of the present method, a spectacle lens area image can be inpainted on the basis of a light reflection area, thereby achieving a better light reflection cancellation effect.
The present disclosure provides an image compression method, an image decompression method, and apparatuses. The image compression method comprises: obtaining a target image, and performing feature extraction on the target image to obtain a first feature map containing a plurality of channels; grouping the channels of the first feature map to obtain a plurality of second feature maps; performing spatial context feature extraction on the second feature maps to determine first spatial redundancy features corresponding to the second feature maps; performing channel context feature extraction on the second feature maps, and determining first channel redundancy features corresponding to the second feature maps; on the basis of the first spatial redundancy features and the first channel redundancy features corresponding to each second feature map, determining compression information corresponding to each second feature map; according to the compression information corresponding to each second feature map, determining first compression data corresponding to the target image, and performing deep compression processing on the basis of the first feature map to determine second compression data corresponding to the target image.
H04N 19/146 - Data rate or code amount at the encoder output
H04N 19/172 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a picture, frame or field
H04N 19/42 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation
G06V 10/44 - Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersectionsConnectivity analysis, e.g. of connected components
60.
IMAGE PROCESSING METHOD AND APPARATUS, AND ELECTRONIC DEVICE, STORAGE MEDIUM, COMPUTER PROGRAM AND COMPUTER PROGRAM PRODUCT
The present disclosure relates to an image processing method and apparatus, and an electronic device, a storage medium, a computer program and a computer program product. The method comprises: extracting initial DCT coefficients of three color components which correspond to a JPEG image to be compressed; respectively arranging the initial DCT coefficients of the three color components according to different frequencies, so as to obtain target DCT coefficients of the three color components; and performing entropy coding on the target DCT coefficients of the three color components, so as to obtain target compressed data corresponding to said JPEG image.
H04N 19/625 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding using discrete cosine transform [DCT]
H04N 19/91 - Entropy coding, e.g. variable length coding [VLC] or arithmetic coding
H04N 19/88 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using pre-processing or post-processing specially adapted for video compression involving rearrangement of data among different coding units, e.g. shuffling, interleaving, scrambling or permutation of pixel data or permutation of transform coefficient data among different blocks
Disclosed in the embodiments of the present application are a sleep wakeup method and a related product. The method comprises: a sleep wakeup apparatus detecting, by means of a microwave module, whether a moving object appears in a first area, wherein the first area belongs to a detection area of the microwave module; when it is detected that no moving object appears in the first area, the sleep wakeup apparatus entering a sleep mode, wherein operation modes of the sleep wakeup apparatus comprise the sleep mode and a wakeup mode, and the power consumption of the sleep wakeup apparatus in the sleep mode is lower than the power consumption thereof in the wakeup mode. In the embodiments of the present application, a sleep wakeup apparatus detects, by means of a microwave module, whether a moving object appears in a first area, and when it is detected that no moving object appears in the first area, the sleep wakeup apparatus enters a sleep mode, such that a sleep wakeup function of the sleep wakeup apparatus can be realized with low power consumption and low costs.
Embodiments of the present application disclose an operation framework of an algorithm applet, a data processing method and device, and a storage medium. The operation framework comprises: an algorithm applet analysis layer used for analyzing an algorithm applet corresponding to a task to obtain an analysis result, the analysis result comprising a configuration file and a script file corresponding to the task, the configuration file being used for configuring operation parameters for the algorithm applet, and the script file being a script program corresponding to a target algorithm for processing the task; an operation pipeline construction layer used for constructing, on the basis of the configuration file and the script file, an operation pipeline for processing the task; and a data stream processing layer used for processing at least one data stream on the basis of the operation pipeline to obtain a processing result corresponding to each data stream.
Provided in the present disclosure are an image generation method and apparatus, an electronic device, and a storage medium. The method comprises: acquiring an original image, wherein the original image includes a target object in an original pose; on the basis of the original image and a target pose parameter, determining a first body posture diagram of the target object in a target pose; determining a first semantic segmentation diagram of the target object in the target pose, wherein the first semantic segmentation diagram is used for indicating a wearing feature of a body part of the target object; and on the basis of the first body posture diagram and the first semantic segmentation diagram, determining a composite image, wherein the composite image comprises the target object, which is in the target pose and has the wearing feature.
The present disclosure relates to a pose detection method, an apparatus, an electronic device, and a storage medium. The method comprises: obtaining at least one first image frame, wherein each among the at least one first image frame comprises at least one two-dimensional image frame of a back face of a target object captured by an image capturing device; and determining pose information of the target object in each of the at least one first image frame according to said at least one first image frame and a preset three-dimensional point cloud of a front face of the target object.
Provided are an image driving method and apparatus, a device and a medium. The method comprises: acquiring a target image and a driving reference image, the target image comprising a body part of a target subject, and the driving reference image comprising a body part of a reference subject presenting a reference action; on the basis of a correspondence between key points of the body part of the target subject and key points of the body part of the reference subject, determining motion information of a plurality of pixel points in the target image, the motion information being used for adjusting an action of the body part of the target subject to be the reference action; and adjusting the plurality of pixel points in the target image according to the motion information so as to obtain a driving effect image, the body part of the target subject in the driving effect image presenting the reference action.
A feature point matching method and apparatus, an electronic device, and a storage medium. A first feature point on a target image, a first coordinate representing the position of the first feature point on the target image, and a first descriptor representing features of the first feature point are determined (S101). Meanwhile, at least one second feature point on a reference image, a second coordinate representing the position of each second feature point on the reference image, and second descriptors representing features of the second feature points are determined (S102). According to the first descriptor and each second descriptor, first screening is performed on the at least one second feature point, so as to obtain at least one candidate feature point (S103); and according to the first coordinate and the second coordinate of each candidate feature point, second screening is performed on the at least one candidate feature point, so as to obtain a target feature point, on the reference image, matching the first feature point (S104).
G06V 10/44 - Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersectionsConnectivity analysis, e.g. of connected components
G06V 10/40 - Extraction of image or video features
67.
POSE PREDICTION METHOD AND APPARATUS, AND DEVICE, MEDIUM, COMPUTER PROGRAM AND COMPUTER PROGRAM PRODUCT
A pose prediction method and apparatus (40), and a device (50), a medium (60), a computer program and a computer program product. The pose prediction method comprises: acquiring several pieces of pose information and several pieces of inertial measurement data of a target object within a first preset time period (S11); determining first certainty representation information of each piece of pose information on the basis of the several pieces of inertial measurement data (S12); and predicting target pose information of the target object within a second preset time period by using the several pieces of pose information, the several pieces of inertial measurement data and the first certainty representation information of each piece of pose information (S13).
G01C 21/16 - NavigationNavigational instruments not provided for in groups by using measurement of speed or acceleration executed aboard the object being navigatedDead reckoning by integrating acceleration or speed, i.e. inertial navigation
An anti-removal target device comprises: an anti-removal detection switch, which is used for detecting whether there is a removal behavior targeting the anti-removal target device, and when detecting that there is a removal behavior, reporting an anti-removal trigger event to a control module; a control module, which is used for determining, under the condition that the anti-removal trigger event is received and the anti-removal target device is in a power-off state, an anti-removal trigger event indication corresponding to the anti-removal trigger event, the anti-removal trigger event indication being used for indicating that there is a removal behavior targeting the anti-removal target device in the power-off state; and a power supply module used for supplying power to the anti-removal detection switch and the control module under the condition that the anti-removal target device is in the power-off state.
G08B 13/14 - Mechanical actuation by lifting or attempted removal of hand-portable articles
G07C 9/00 - Individual registration on entry or exit
G08B 7/06 - Signalling systems according to more than one of groups Personal calling systems according to more than one of groups using electric transmission
69.
IMAGE DETECTION AND TRAINING METHOD, RELATED APPARATUS, DEVICE, MEDIUM, AND PROGRAM PRODUCT
The present application discloses an image detection and training method, a related apparatus, a device, a medium, and a program product. A training method for an image detection model comprises: obtaining a sample medical image and reference information thereof, wherein the reference information comprises a first distance corresponding to a target pixel point, the first distance represents the actual distance from the target pixel point to a first reference position, and the first reference position represents the center position of a sample object to which the target pixel point belongs; processing the sample medical image by using an image detection model so as to obtain a first processing result, wherein the first processing result comprises a second distance corresponding to the target pixel point, and the second distance represents a predicted distance from the target pixel point to the first reference position; and obtaining difference information on the basis of the first distance and the second distance corresponding to the target pixel point, and adjusting network parameters of the image detection model on the basis of the difference information. The above solution may accurately distinguish between and detect different objects even when said objects are very close to each other or even adhered to each other.
Provided in the present disclosure are a traffic statistics method and apparatus, and a computer device and a storage medium. The method comprises: acquiring at least one target image corresponding to a target region within a preset time period; determining, on the basis of the at least one target image, a first number of objects that corresponds to each position in the target region, and a staying time of each object at the corresponding position; for each position in the target region, obtaining, on the basis of the first number of objects that corresponds to the position and the staying time of each object at the position, a second number of objects that corresponds to the position; and generating heat map statistical information of object traffic of the target region on the basis of the second number corresponding to each position in the target region. By means of the embodiments of the present disclosure, a second number of objects that corresponds to each position in a target region can be flexibly and accurately determined, and accurate heat map statistical information, which corresponds to object traffic, of the target region is generated.
The present invention relates to a crowd positioning method and apparatus, an electronic device, and a storage medium. The method comprises: performing positioning of human body key points of a crowd image to obtain an initial positioning map corresponding to the crowd image, wherein the initial positioning map is used for indicating positions of initial human body key points comprised in the crowd image; on the basis of the positions of the initial human body key points in the crowd image, determining a target neighborhood corresponding to the initial human body key points; and on the basis of the target neighborhood corresponding to the initial human body key points, filtering the initial positioning map to obtain a target positioning map, wherein the target positioning map is used for indicating positions of target human body key points comprised in the crowd image. According to embodiments of the present invention, a false detection probability that a same human body corresponds to a plurality of initial human body key points can be reduced, and the target positioning map having relatively high accuracy is obtained.
Disclosed in the embodiments of the present disclosure are a board-game playing explanation scheme generation method and apparatus, and an electronic device, a storage medium and a program product. The method comprises: during a board-game playing process, on the basis of a game of a board-game play, determining candidate moves of an attacking side in each deduction round of a preset number of deduction rounds, on the basis of the candidate moves, estimating recommended moves of the attacking side in the current round of the board-game play, and performing a play on the basis of the recommended moves, so as to generate a current game of a board-game play; when there is no winner in the current game of the board-game play, on the basis of the current game of the board-game play, estimating recommended moves of the attacking side in the next round of the board-game play, performing a play according to the recommended moves of the attacking side in the next round of the board-game play, so as to generate a next game of a board-game play until there is a winner in a generated game of a board-game play, and generating explanation text of each candidate move and each recommended move; and on the basis of the explanation text, generating a board-game playing explanation scheme of the board-game playing process.
The present disclosure relates to a painting display method and apparatus, an electronic device, a storage medium, and a program product. The method comprises: in response to the fact that a real-scene image acquired by the electronic device comprises a target painting, displaying, in a target painting area displayed by the electronic device, a virtual character corresponding to a target user, wherein the target painting area represents an area where the target painting is located in the real-scene image displayed by the electronic device; and in response to the fact that a distance between the virtual character and any target object in the target painting area is smaller than or equal to a preset distance, displaying virtual information corresponding to the target object.
Provided in the present disclosure are a model presentation method and apparatus, and an electronic device and a storage medium. The method comprises: acquiring a projection image collected by a projection camera, and a thumbnail corresponding to each three-dimensional model comprised in the projection image; determining region information of an image superposition region, which is set in the projection image; on the basis of projection information of the three-dimensional models in the projection image and the region information, determining superposition information of the thumbnails corresponding to the three-dimensional models, wherein the superposition information comprises: superposition positions and/or presentation sizes on the image superposition region; and according to the superposition information of the thumbnails, superposing, onto the image superposition region of the projection image, the thumbnails corresponding to the three-dimensional models, so as to generate a target presented image.
Provided in the embodiments of the present disclosure are a map construction method and apparatus, a pose determination method and apparatus, and a device and a computer program product. The map construction method comprises: acquiring several target image pairs, which are obtained by means of photographing a target environment, wherein each target image pair comprises two scene images; on the basis of the several target image pairs, determining a pose change between the two scene images in each target image pair; and obtaining location information of several map points in the target environment by using the pose change and reference positioning information of the scene images, wherein the reference positioning information is obtained by means of performing positioning on the scene images by using a preset positioning mode; and the map points are spatial points, which correspond to image points in the several target image pairs, in the target environment.
The present disclosure provides a method and apparatus for special effect rendering, a device, and a storage medium. The method comprises: obtaining a target image, the target image comprising a target object; obtaining a 3D object model, the 3D object model comprising a target area; and on the basis of depth information of the 3D object model and color information of a non-target area except the target area on the 3D object model, rendering the 3D object model on the target image to obtain a rendering result image, wherein the target object is displayed in the target area.
An experiment presentation method and apparatus, and an electronic device, a storage medium and a program product. The method comprises: in response to a production request from a first user for an experiment task to be produced, acquiring a meta-information file that matches said experiment task, wherein the meta-information file is used for defining an experiment structure to which an experiment of the same type belongs, which experiment corresponds to said experiment task (S101); parsing the meta-information file, and determining at least one interface module for performing experiment interface presentation (S102); and combining the at least one interface module, so as to obtain a presentation result for a produced experiment task (S103). In the method, experiment structures, to which various types of experiments belong, can be predefined on the basis of a meta-information file, and the production of related experiment tasks can be realized during the whole experimental process on the basis of an interface module obtained by analyzing the experiment structures; and the method is easy to operate and has a better applicability, and thus can be widely applied to teaching and research work.
The present invention provides an online course resource use method and apparatus, a device, a medium, and a program product. The method comprises: determining a target online course selected in multiple online courses by a user; determining, according to a corresponding course purchase record of the user, a user use permission corresponding to the target online course, the user use permission comprising an official use permission or a trial permission; and displaying a course detail page of the target online course according to the determined user use permission, wherein for different user use permissions, operation functions supported by the course detail page are different.
Provided in the present disclosure are a loop instruction processing method and apparatus, and a chip, an electronic device, and a storage medium. The loop instruction processing method comprises: acquiring a multi-layer loop instruction, wherein the multi-layer loop instruction includes a plurality of layers of a nested loop instruction; acquiring, from a parameter register of a processor, a loop parameter of each layer of loop instruction in the multi-layer loop instruction; during an execution process of each layer of loop instruction, determining execution logic of a loop body in the layer of loop instruction on the basis of state information of the layer of loop instruction and the loop parameter of the layer of loop instruction, wherein the state information is used for indicating a real-time execution state of the layer of loop instruction; and controlling the execution of the loop body in the layer of loop instruction on the basis of the execution logic.
Provided in the present disclosure are an instruction processing apparatus and method, a chip, a computer device, and a storage medium. The method comprises: acquiring an instruction to be processed, and parsing said instruction, so as to obtain a parsing result; when, on the basis of the parsing result, the mode in which a register file is addressed by means of said instruction is determined to be an indirect address mode, determining an index register in the register file, wherein index information in the index register is used for indicating an address where a target operand for executing said instruction is stored in the register file; acquiring the index information in the index register, and determining a target register matching the index information in the register file; and executing said instruction on the basis of the target operand stored in the target register.
The present disclosure relates to a counting method and apparatus, an electronic device, a storage medium and a computer program product. The method comprises: performing feature extraction on an input image to obtain a first feature map and a second feature map; extracting an object feature of an object to be counted from the first feature map according to position information of at least one object to be counted annotated in the input image; determining distribution position information of at least two objects to be counted in the first feature map according to the object feature; determining a feature distribution map of said object from the second feature map by utilizing the distribution position information; and determining the total number of objects to be counted on the basis of the feature distribution map. According to embodiments of the present disclosure, the accuracy of visual counting can be improved.
Embodiments of the present disclosure provide a data processing method and apparatus, a computer device, a storage medium, and a computer program product. The method comprises: obtaining first position information of at least one object in a first coordinate system; on the basis of a first conversion parameter, converting the first position information of each object in the first coordinate system into second position information in a second coordinate system; determining target operation information on the basis of the second position information of each object and a preset operation rule, the target operation information comprising a target operation object, a target operation instruction, and target position information in the second coordinate system; and executing the target operation instruction on the basis of the target operation object and the target position information.
Disclosed in the present application are an adjustment circuit and an electronic device. The adjustment circuit comprises: a current limiting circuit, which is used for receiving a power source provided by an external power source circuit, and performing a configurable setting on the maximum output current of the power source; an energy storage circuit, which is coupled to the current limiting circuit, and is used for receiving a current-limited power source output by the current limiting circuit, so as to store energy; and a power amplification circuit, which is coupled to the current limiting circuit, the energy storage circuit and an external load, wherein the power amplification circuit is used for receiving an externally input audio signal, acquiring a state value of the load to receive, according to the state value, the power source, the current of which is limited by the current limiting circuit, or to receive, according to the state value, an energy storage power source provided by the energy storage circuit and the power source, the current of which is limited by the current limiting circuit, so as to adjust the audio signal, and outputting the adjusted audio signal to the load. Therefore, the adjustment circuit in the present application can effectively reduce the peak-to-average ratio of the power consumption of a system, such that the peak power consumption of the system is effectively reduced when indexes such as the degree of distortion of an output of a power amplifier are met.
H04B 5/02 - Near-field transmission systems, e.g. inductive loop type using transceiver
H02J 7/00 - Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
H02J 7/02 - Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries for charging batteries from AC mains by converters
84.
FACIAL TRACKING RECOGNITION METHOD AND APPARATUS, AND ELECTRONIC DEVICE, MEDIUM, AND PROGRAM PRODUCT
A facial tracking recognition method and apparatus, and an electronic device, a medium, and a program product. The method comprises: determining an ith frame in a video frame sequence to be an initial key frame (S10); performing global facial detection on the ith frame, so as to obtain at least one first detection box therein which represents a facial position, and facial feature information of a facial region in each first detection box; adjusting the at least one first detection box in the ith frame, so as to obtain second detection boxes of an (i+1)th frame (S30); acquiring a detection region in each second detection box of the (i+1)th frame, and performing local facial detection on same, so as to obtain at least one first detection box therein which represents a facial position, and facial feature information of a facial region in each first detection box; and comparing the facial feature information in the ith frame with the facial feature information in the (i+1)th frame, so as to obtain identical faces (S60). By means of the method, detection boxes in the previous frame among consecutive video frames are adjusted, so as to determine a facial detection range in the current frame, such that the computing power required during a facial detection process can be reduced, thereby improving the facial tracking recognition efficiency.
Embodiments of the present disclosure disclose a human-machine chess playing method and apparatus, a device, a storage medium, and a computer program product. The method comprises: according to an acquired current chess game and a guidance level corresponding to a first party, determining a piece placement strategy corresponding to the first party; guiding the first party to play the current chess game by means of the piece placement strategy; in response to a piece placement event of the first party in a current turn, and according to a first piece placement position corresponding to the piece placement event, determining a second piece placement position corresponding to a second party in the current turn; and adjusting the guidance level according to the piece placement strategy, the first piece placement position and the second piece placement position.
The present invention relates to an image processing method and apparatus, an electronic device, and a storage medium. The processing method comprises: acquiring a target image, the target image comprising a target object; detecting the face category to which the target object belongs in the target image; determining a correction parameter of a face area of the target object on the basis of the face category; and adjusting the face area in the target image on the basis of the correction parameter.
Provided in the present disclosure are an instruction processing method and apparatus, a chip, an electronic device, and a storage medium. The instruction processing method comprises: acquiring a first computation instruction to be processed; determining an addressing mode of first address information of first access data corresponding to the first computation instruction, wherein the first access data comprises an operand which is used for executing the first computation instruction and/or a computation result of the operand; when the addressing mode is an address accumulation mode, acquiring accumulated step size information on the basis of the first computation instruction, and acquiring second address information of second access data corresponding to a second computation instruction, wherein the second computation instruction is the instruction preceding the first computation instruction; and on the basis of the second address information and the accumulated step size information, determining the first address information, and on the basis of the first address information, executing the first computation instruction.
The present disclosure relates to an abnormal event detection method and apparatus, a computer device, a storage medium, a computer program, and a computer program product. The method comprises: acquiring at least two image sequences, wherein each of the image sequences comprises at least one image frame; performing division at at least two scales on each image sequence to obtain image block sets made up of image blocks at a same position in all image frames at a same scale; determining a correlation characteristic between each image sequence on the basis of the image block set of each image sequence; and determining a target image sequence where an abnormal event is present among the at least two image sequences according to the correlation characteristic between each image sequence. The accuracy of abnormal event detection can be improved by means of the present method.
Embodiments of the present disclosure provide model training and image recognition methods and apparatuses, a device, a storage medium and a computer program product. The model training method comprises: acquiring a first image sample containing a first object; performing feature extraction on the first image sample using a first network of a first model to be trained, to obtain a first feature of the first object; updating the first feature on the basis of a second feature of at least one second object, using a second network of the first model, to obtain a first target feature corresponding to the first feature, a degree of similarity between each second object and the first object being not less than a first threshold; determining a target loss value on the basis of the first target feature; and updating a model parameter of the first model at least once on the basis of the target loss value, to obtain a trained first model.
The present disclosure provides a commodity template generation method and apparatus, an electronic device, a storage medium, and a computer program product. The method comprises: in response to a configuration instruction for a created initial commodity template, selecting, from a candidate commodity library, a target commodity to be added to the initial commodity template; determining a degree of similarity between the target commodity and each of existing commodities in the initial commodity template; and in response to the fact that the degree of similarity is less than a preset threshold, updating the initial commodity template according to the target commodity to obtain a target commodity template, the target commodity template being configured to guide an operator to execute a target operation on a target vending terminal.
G06V 10/75 - Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video featuresCoarse-fine approaches, e.g. multi-scale approachesImage or video pattern matchingProximity measures in feature spaces using context analysisSelection of dictionaries
G06V 10/74 - Image or video pattern matchingProximity measures in feature spaces
G07F 9/00 - Details other than those peculiar to special kinds or types of apparatus
G07F 9/02 - Devices for alarm or indication, e.g. when emptyAdvertising arrangements in coin-freed apparatus
G06Q 10/08 - Logistics, e.g. warehousing, loading or distributionInventory or stock management
91.
IMAGE PROCESSING METHOD AND APPARATUS, AND ELECTRONIC DEVICE, STORAGE MEDIUM AND COMPUTER PROGRAM PRODUCT
The present disclosure relates to an image processing method and apparatus, and an electronic device, a storage medium and a computer program product. The method comprises: determining a first mask image and a second mask image of a facial image, wherein the first mask image and the second mask image are used for representing a first part area and a second part area, respectively, in the facial image, and the second part area is in the first part area; acquiring lip processing information comprising area information and beautification parameter information; adjusting the range of the second part area in the second mask image according to the area information and the first mask image; and beautifying the facial image according to the beautification parameter information and the adjusted second mask image, so as to obtain a target image in which the color of the second part area is adjusted.
Disclosed in the embodiments of the present application are an action generation method and a related apparatus, and an electronic device, a storage medium and a program. The action generation method comprises: acquiring a first feature representation respectively representing a plurality of individuals in a plurality of action frames, and acquiring a second feature representation respectively representing the plurality of individuals with respect to a target action category; performing relationship modeling on the basis of the first feature representation and the second feature representation, so as to obtain a fused feature representation of the individuals in the action frames, wherein the type of relationship modeling is related to a first total number of the plurality of individuals; and performing action mapping on the basis of the fused feature representation, so as to obtain an action sequence of the plurality of individuals with respect to the target action category, wherein the action sequence comprises a plurality of action frames, and the action frames include action representations of the individuals.
Disclosed in the present application are a training method and apparatus for a target detection model, a corresponding detection method and apparatus for a target detection model, and a device, a storage medium and a computer program product. The training method comprises: acquiring a sample medical image which includes a preset organ, wherein the sample medical image is annotated with an annotation result of at least one target which is located on the preset organ, and the annotation result comprises an actual area where the target is located; by using a target detection model, respectively matching at least one first candidate area for the at least one target according to a matching sequence, and obtaining a final prediction result with regard to the target on the basis of the first candidate area, wherein the matching sequence is determined on the basis of the size of the actual area where the at least one target is located; and adjusting the parameters of the target detection model by using the final prediction result and the annotation result. By means of the method, a recall rate of a target detection model during training can be improved.
Disclosed in embodiments of the present disclosure are a remote sensing data processing method and apparatus, a device and a storage medium. The method comprises: obtaining original remote sensing data; receiving an interpretation instruction for the original remote sensing data, and generating an interpretation task corresponding to the original remote sensing data, wherein the interpretation task is used for generating an interpretation result corresponding to the original remote sensing data; and generating a corresponding remote sensing map on the basis of the original remote sensing data and the interpretation result.
G06F 16/587 - 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
95.
OBJECT RECOGNITION METHOD, NETWORK TRAINING METHOD AND APPARATUS, DEVICE, MEDIUM, AND PROGRAM
Disclosed in embodiments of the present invention are an object recognition method, a network training method and apparatus, a device, a medium, and a program. The object recognition method comprises: acquiring a video frame to be recognized comprising a target object of a screen; said video frame being any video frame in a video stream of the target object; determining an initial attitude sequence of the target object on the basis of said video frame and a historical video frame of said video frame in the video stream; performing probability mapping on the initial attitude sequence to obtain a target attitude sequence of the target object in said video frame; performing feature conversion on the target attitude sequence in space and time to obtain an attitude feature trajectory of the target object in said video frame; and determining a behavior state of the target object in said video frame on the basis of the attitude feature trajectory.
Embodiments of the present invention provide a model management method, a model management service, a client end, a storage medium, and a program. The method comprises: if a model management service meets a preset synchronization condition, synchronizing a target model file and target model meta-information of a target model from an object storage service; and in response to a load request of a view resolution service, sending the target model file and the target model meta-information to the view resolution service.
A payment device. The payment device comprises: a base (1), a cavity of the base (1) being provided with an auxiliary board (16), the auxiliary board (16) being provided with an interface (13), and the base (1) being provided with an opening (14) for exposing the interface (13); a stand (2), connected to the base (1); a first screen (3), connected to the stand (2), a main board (34) being provided in a case (35) of the first screen (3), and the main board (34) being electrically connected to the auxiliary board (16); and a second screen (4), connected to the stand (2).
F16M 11/04 - Means for attachment of apparatusMeans allowing adjustment of the apparatus relatively to the stand
F16M 11/10 - Means for attachment of apparatusMeans allowing adjustment of the apparatus relatively to the stand allowing pivoting around a horizontal axis
G06Q 20/40 - Authorisation, e.g. identification of payer or payee, verification of customer or shop credentialsReview and approval of payers, e.g. check of credit lines or negative lists
G06F 3/041 - Digitisers, e.g. for touch screens or touch pads, characterised by the transducing means
G06V 40/16 - Human faces, e.g. facial parts, sketches or expressions
G06V 40/40 - Spoof detection, e.g. liveness detection
G06Q 20/42 - Confirmation, e.g. check or permission by the legal debtor of payment
The present disclosure relates to a method for displaying an augmented reality (AR) scene. The method comprises: in response to an identification object being recognized in a first image of the current scene, and according to first pose data of a virtual object, in a world coordinate system, corresponding to the identification object, determining second pose data of the virtual object in a screen coordinate system of an AR device; displaying the virtual object on the AR device on the basis of the second pose data; and during the process of displaying the virtual object, in response to the identification object not being recognized in a second image of the current scene, displaying the virtual object on a screen of the AR device on the basis of preset target pose data in the screen coordinate system.
The present disclosure relates to an AR scene display method, comprising: in response to identification of an identification object from a current scene image, displaying a virtual object on an AR device on the basis of display pose data determined according to first relative pose data of the AR device and the virtual object; in the display process of the virtual object, determining target display data of the virtual object according to second relative pose data of the virtual object and the identification object; and displaying the virtual object on the AR device according to the display pose data and the target display data.
G06T 19/00 - Manipulating 3D models or images for computer graphics
100.
TRAJECTORY POINT SCREENING METHOD AND APPARATUS, TRAJECTORY GENERATION METHOD AND APPARATUS, ELECTRONIC DEVICE, STORAGE MEDIUM AND COMPUTER PROGRAM PRODUCT
Disclosed in the embodiments of the present disclosure are a trajectory point screening method and apparatus, a trajectory generation method and apparatus, a device, a storage medium and a computer program product. The trajectory point screening method comprises: acquiring M trajectory points of a real driving trajectory; acquiring a current key trajectory point from the M trajectory points, and determining a current basic slope on the basis of the current key trajectory point; and when the current trajectory point is not the Mth trajectory point, determining, on the basis of the current basic slope, a slope between a previous trajectory point and the current trajectory point in the M trajectory points, and a slope between the key trajectory point and the current trajectory point, a next key trajectory point from the M trajectory points and a next basic slope until trajectory points to be screened in the M trajectory points are traversed, so as to obtain N-1 key trajectory points, the N-1 key trajectory points and the Mth trajectory point being used for generating a simulation trajectory.
patents
