323 results about "Object Attribute" patented technology

The invention relates to the technical field of robots, and discloses a task planning method and device for a robot, equipment and a medium, and the method comprises the steps: building a scene relation graph model containing environment space topology and object attributes; inputting the received natural language instruction into the first language model, and outputting a structured task target; on the basis of the structured task target and the real-time environment perception data, auxiliary information is generated through a second language model, and the scene relation graph model is updated; inputting the updated scene relation graph model into a reinforcement learning model, and outputting an action decision sequence; and the target robot is controlled to execute the action decision sequence, and reinforcement learning strategies of the scene relation graph model and the reinforcement learning model are adjusted according to the environment feedback information. According to the invention, efficient, flexible and accurate task planning of the robot in a complex dynamic scene can be realized, and the efficiency and adaptability of task planning of the robot are improved.

The invention discloses a man-machine interaction method and system based on an intelligent sensor, belongs to the technical field of robots, solves the problem of misoperation caused by abnormal sensor data to a certain extent, and enables the system to still run stably through quality credibility evaluation and a dynamic weight distribution mechanism when part of sensors fail. Meanwhile, based on a dynamic safety boundary strategy of scene semantic understanding, the robot can automatically adjust motion constraints according to object attributes and spatial relations, and collaborative optimization of safety and operation efficiency is achieved in complex tasks.

The invention provides a control method and system of a sweeping robot, and the method comprises the steps: obtaining image data, three-dimensional space data and sound data of a surrounding environment, and extracting object attribute features, space features and sound attribute features in the environment; fusing the object attribute feature, the spatial feature and the sound attribute feature to obtain an identification result; based on an identification result, executing hierarchical scene adaptive planning and executing a cleaning task by adopting a self-adaptive variational cleaning strategy; after the cleaning task is completed, second image data of the cleaned area are collected; recognizing residual stains on the ground according to the second image data and a preset cleaning effect recognition model; and if the area of the residual stain exceeds a preset area threshold value, triggering a re-cleaning program, and automatically switching a corresponding cleaning mode according to the type of the residual stain. According to the scheme, when the sweeping robot encounters the obstacles in the sweeping area, the sweeping robot can intelligently take accurate countermeasures, and an excellent sweeping effect is achieved.

The invention relates to a method and a device for generating a video with a body and electronic equipment. The method comprises the following steps: analyzing a task instruction and initial environment observation data, generating a key operation step sequence and an associated object set thereof, and constructing a three-dimensional physical constraint graph; generating an initial action video sequence based on the coding condition; in response to single-frame action execution, calculating a spatial error index, and if the object contact distance deviation, the trajectory collision probability or the physical constraint violation value exceeds a preset threshold, triggering the space-time diffusion model to generate a correction frame based on the current physical constraint graph; in response to the completion of the key operation steps, verifying the semantic matching degree of the generated result and the task target, and if the key object attribute loss or the operation logic error is detected, regenerating the task operation step sequence and the physical constraint graph; and updating the object position, the distance threshold and the motion feasible region in the physical constraint graph, and inputting the space-time diffusion model. According to the method, the problems of physical consistency deficiency, error accumulation and semantic understanding splitting are solved.

The invention relates to the technical field of data processing, and discloses a full-life-cycle-oriented engineering technology data governance method and a full-life-cycle-oriented engineering technology data governance system. The method comprises the steps that project object attributes are divided into an anchor point attribute set and an extended attribute set, a locking mark is added to the anchor point attribute set, the extended attribute set is layered according to stages to obtain a data object meta-model, a dependency relationship mapping table is established according to the data object meta-model to achieve automatic verification when the attributes are changed, and the attribute change is completed. And establishing a spatial topological graph and a process graph to diagnose inconsistent defects, establishing a weighted graph to quantitatively analyze change influences, evaluating data quality according to scene types, and intelligently complementing missing attributes. According to the method, the integrity and consistency of cross-stage transmission of engineering data, the real-time performance and accuracy of multi-professional data collaboration, the depth and automation degree of data quality diagnosis, the scientificity and comprehensiveness of change impact analysis and the flexibility and efficiency of data utilization are improved.

Methods and systems for prompting a large language model (LLM) to generate a description of an object with indications of any unsubstantiated information are disclosed. A prompt is generated to a LLM to generate a description of an object, where the prompt includes one or more object attributes to include in the generated description. The prompt also includes an instruction for the LLM to annotate any portions of the generated description that are, involve, and / or include unsubstantiated information according to a defined format. The prompt is provided to the LLM and the generated description is received. The generated description is parsed to identify, based on the defined format, one or more annotated portions indicating unsubstantiated information. The generated description is presented for display via a user device.

A computer-implemented method is disclosed. The method includes: receiving, via a first user interface, a selection associated with an object; determining a first set of object attributes based on the selection; presenting, via a second user interface, a text prompt for a user to identify a subset of the first set of object attributes; receiving, via the second user interface, an indication of one or more preferred object attributes of the identified subset; and updating the first user interface to display content relating to objects associated with the one or more preferred object attributes.

The invention discloses a multi-object arrangement method and system for a mobile operation robot, and the method comprises the steps: obtaining an object image of the robot at a current visual angle, and extracting the attribute information of an object; constructing a bearing relation matrix and a semantic scene graph; the bearing relation matrix is converted into a directed acyclic graph, an edge set serves as a state space in reinforcement learning, an action space is defined to select one of unselected edges, the depth of the edge set serves as the basis of current grouping, and objects are grouped based on a high-level reinforcement learning model; obtaining a stacking execution sequence of the objects in the group based on the semantic scene graph; and the grabbing posture of the current to-be-operated object is obtained through prediction of the grabbing network, the object is grabbed according to the grabbing posture, and the grabbed object is stacked according to the planned stacking sequence. According to the robot, efficient action decision making is achieved in the grabbing, stacking and placing processes, and the robot has the capacity of carrying multiple objects in different places and orderly placing the multiple objects in a complex environment.

The invention discloses a multi-target constraint force control parameter determination method and a control system for a machine dexterous hand, and the method comprises the steps: obtaining multi-mode state data, determining a current operation stage and an object attribute, constructing a candidate force control parameter set, carrying out the multi-target optimization through employing an NSGA-II algorithm under multiple constraints, obtaining a Pareto optimal solution set, and carrying out the multi-target constraint force control of the machine dexterous hand. And after the target force control parameter group is determined, performing safety projection correction on original control output, and controlling the dexterous hand to execute operation. According to the scheme, through closed-loop collaborative design of stage recognition, parameter optimization, safety correction and online reconstruction, contact safety, grabbing stability, response performance and energy consumption control are considered, the grabbing success rate is increased to 95% or above from 75%, the contact peak force is reduced by 36%, and the control reliability of multi-finger collaborative operation is improved.

Techniques are discussed herein for generating and evaluating driving simulations based on synthetic scenarios. Simulated objects may be controlled based on parameters determining the attributes and behaviors of the objects, and scenarios may be synthetically modified by changing the parameters for a simulated object. For a driving scenario with a synthetic simulated object, a simulation system may analyze driving log data to determine a probability or prevalence associated with the driving scenario. In various examples, the simulation system may determine marginal density estimates for individual attributes of the simulated object, as well as a cumulative distribution function modeling the dependence between the attributes. The joint probability distribution determined for the synthetic simulated object can be used for evaluating the efficacy of the simulation and the performance of the simulated vehicle controllers.

The invention provides an information display method and device, equipment, a storage medium and a product, and relates to the technical field of computers. The method comprises the steps that in a first device, a first object information set is displayed in a first preset page, the first object information set comprises first object information of a preset object in a preset object set, the preset object set is associated with a target live video interface, and target prompt information is displayed at the position adjacent to the first object information of the first target preset object, the business data of the first target preset object meets a preset requirement, the target prompt information is used for prompting that the first target preset object is set as a preset object with a target object attribute, and second object information of the preset object with the target object attribute contains a target label and is displayed at a target position; the second object information set is displayed on a second preset page associated with the target live video interface in the second device. By adopting the technical scheme, the setting efficiency of the target object attribute of the preset object can be improved.

The invention discloses an object identification method and device, a storage medium and electronic equipment. The method comprises the steps that a to-be-recognized image is acquired, and the to-be-recognized image is used for indicating object information in a detection area; the to-be-recognized image is input into a target recognition model to obtain a target recognition result, the target recognition model comprises a defogging branch and a super-resolution branch, the super-resolution branch is used for amplifying the image resolution of the to-be-recognized image to determine the super-resolution feature of the to-be-recognized image, the defogging branch is used for adjusting the local contrast of the to-be-recognized image, and the super-resolution feature of the to-be-recognized image is obtained. The target recognition result is determined according to the super-resolution feature, the defogging feature and the original image feature of the to-be-recognized image, and the target recognition result is used for indicating the object position and the object attribute of the target object in the detection area. The technical problem that the object recognition accuracy is low due to the fact that the image collected in the to-be-detected area is not clear enough is solved.

Embodiments of the present disclosure provide large language models (LLMs) and large multimodal models (LMMs) for managing imaging data. An example method includes extracting a database schema and a data dictionary associated with a medical imaging database that maintains medical images and non-image data associated with a plurality of clinical studies; extracting object attributes associated with the medical images that are descriptive of the medical images and corresponding studies; and training an LLM to generate search queries from natural language requests based on a training dataset that includes: the database schema, the data dictionary, the object attributes, a prompt template including partial instructions for the LLM, and a plurality of example natural language requests and corresponding expected search queries. The trained LLM can then be used to generate a first search query from a first natural language request.

The invention provides a deep processing method and system for a digital delivery result, and relates to the technical field of data processing. A delivery result file set is disassembled and recombined into a unified object main data set and a three-dimensional entity set; each object attribute record and each object relation record have a unified object identifier, a source identifier and evidence positioning information at the same time, and multiple sets of coding calibers of the same object can be gathered to the same unified object identifier; refined model replacement is executed by taking a unified object identifier as an anchor point in three-dimensional treatment, continuity of a composition relationship and a connection relationship is kept, and after project handover and commissioning, online and field inconsistency found by operation and maintenance personnel is converted into a revision chain capable of being rechecked, judged and audited, so that the operation and maintenance efficiency is improved. Therefore, the problems that the unstructured data is difficult to reuse, the coding calibers are different, the object relationship is easy to lose and the data in the operation and maintenance period is not credible are solved.

Embodiments disclosing a controller for controlling a robot to perform a task are provided. The task is performed in an environment that is represented by an input image. The controller causes segmenting of an object in the input image. A confidence level of segmentation is updated by comparing the segmented object with constrained affined transformations of a template of the object. The constrained affine transformations are based on constraints indicative of a property of the object. The property of the object and the updated confidence level of segmentation are then used for performing the task.

The embodiment of the invention provides a method and device for identifying logistics attributes, electronic equipment and a storage medium. The method comprises the steps that object information of a logistics object to be recognized is acquired, and the object information is used for representing object attribute information of the logistics object; a plurality of logistics categories of the logistics object are determined according to the object information, and the logistics categories represent corresponding categories of the logistics object in a category system constructed based on logistics attributes; according to a preset mapping relation, candidate logistics attributes corresponding to all the logistics categories are determined, target logistics attributes corresponding to the logistics objects are determined based on the priorities of all the candidate logistics attributes, and the preset mapping relation comprises the mapping relation between the logistics categories of the logistics objects and the candidate logistics attributes. According to the embodiment of the invention, the identification of the logistics attributes is converted into the identification of the logistics categories, the influence of the change of factors such as logistics policies on the identification result of the logistics attributes is avoided, and the accuracy of the identification of the logistics attributes is improved.

A computer-implemented is disclosed. The method includes: obtaining a composite image depicting an identifiable foreground object; performing image segmentation to isolate a foreground object from the composite image, the image segmentation yielding a segmented composite image; determining object attributes of the foreground object based on image analysis of the segmented composite image; obtaining at least one prompt for a text-to-image model based on the object attributes of the foreground object, the at least one prompt defining an intended replacement background for the composite image; providing, to the text-to-image model, instructions to generate a replacement background image for the composite image based on the at least one prompt; receiving, from the text-to-image model, a replacement background image for the composite image; and generating a new composite image, the generating including compositing portions of the composite image corresponding to the foreground object with the replacement background image.

The invention discloses a three-dimensional modeling method based on voice interaction and a large language model, and relates to the technical field of computer intelligent modeling. According to the method, on the basis of dialogue context management, a closed-loop'voice / text input-modeling script generation-execution monitoring-error feedback and repair 'process is constructed based on a Rhinoscriptsyntab library in combination with RAG (Retrieval Enhanced Generation) and an automatic error correction mechanism, so that a three-dimensional model can be incrementally modified in multiple rounds, and object attributes can be tracked; therefore, flexible response to complex modeling requirements is realized in the Rhino environment. According to the method, a designer can quickly generate and edit the three-dimensional model through a natural language instruction, so that the design efficiency is remarkably improved, the cognitive burden of software operation is reduced, and the continuity of creative thinking is kept.

The present disclosure is directed to systems and methods for performing object detection and pose estimation in 3D from 2D images. Object detection can be performed by a machine-learned model configured to determine various object properties. Implementations according to the disclosure can use these properties to estimate object pose and size.

The invention relates to an inspection data management method and system for an unmanned aerial vehicle, and relates to the field of data management, and the method comprises the steps: obtaining a first field image and first point cloud data collected by the unmanned aerial vehicle at an inspection starting moment, and a second field image and second point cloud data collected after a preset time interval; extracting a first object and a second object, and obtaining a first object attribute of the first object and a second object attribute of the second object; according to the first object attribute and the second object attribute, obtaining an inspection scheme of the unmanned aerial vehicle; the unmanned aerial vehicle executes the inspection scheme and collects inspection data; and packaging the inspection data according to the object category and the timestamp, and uploading the inspection data to a data management terminal. In this way, the utilization efficiency and the data acquisition quality of the unmanned aerial vehicle are improved, and then accurate management of inspection data is achieved.

The invention relates to the technical field of data processing, in particular to a GPU architecture conversion method and device. The method comprises the steps that assembly code input is received, ISA is analyzed into lexical units containing instruction elements according to input content, then detection of grammar rules and semantic logic is conducted on the lexical units based on grammar standards, and instruction objects carrying complete instruction attributes are generated; converting the instruction object into a corresponding LLVM function according to the attribute of the instruction object, generating an initial intermediate representation code, and finally compiling the optimized intermediate representation code into an executable machine code according to the target GPU hardware type. Through the embodiment of the invention, efficient conversion from the GPU assembly code to the multi-architecture machine code is realized, and the problem of performance loss caused by inaccurate grammar detection and insufficient optimization in traditional GPU development can be solved.

The invention provides an author page value prediction model training method and device, equipment and a storage medium. The invention relates to the technical field of artificial intelligence, in particular to the technical fields of deep learning, large models, natural language processing, data analysis and the like, and can be used for application scenes such as resource recall, resource scoring, resource recommendation and the like. The specific scheme is as follows: obtaining an interaction log of a target object and a target resource; determining object attribute features of the target object, resource attribute features of the target resource and related interaction features of the target object to an author page of the target resource based on the interaction log; generating a multi-dimensional feature set based on the object attribute features, the resource attribute features and the related interaction features; and taking the multi-dimensional feature set as input of a to-be-trained model, taking the author page consumption classification, the author page consumption duration and the author page consumption step length as prediction targets of the to-be-trained model, and training the to-be-trained model to obtain an author page value prediction model.

A development system provides an editor and a simulator between which changes are bi-directionally synchronized. As the simulator runs the application logic, objects may be spawned and destroyed. Object properties (e.g., colors, positions, behaviors, etc.) can also change as time passes. The developer can see these changes in the editor view in real time. The developer can make changes in the editor view that persist and are synchronized in real time to the simulator.

The invention discloses an event ontology mode construction method and device under the assistance of a large model, and belongs to the technical field of ontology construction and evolution.The event ontology mode construction method comprises the steps that a large language model is used for precisely screening corpus metadata and automatically extracting event elements; the method comprises the following steps: converting an event quintuple into semantic, time sequence and relation vectors through a deep semantic representation model, dynamically fusing time sequence and relation characteristics by taking the semantic vector as a dominant, and iteratively merging the dynamic event vectors by adopting a time sequence-relation enhanced hierarchical clustering algorithm to construct a clustering tree with a father-child hierarchical relation; generating general Chinese class names for non-leaf nodes through a large language model, and forming an event ontology skeleton which is easy to understand; and finally, the ontology skeleton is mapped into OWL ontology classes, individuals and object attributes conforming to the RDF / OWL specification, a standardized ontology file is generated, automatic construction and output of the event ontology mode are achieved, and the automation degree and normalization of event ontology construction are improved.

The application discloses a terminal access permission analysis method and system based on a zero-trust ABAC model, and the method comprises the following steps: a user terminal sends a resource access request; a policy enforcement point (PEP) receives the resource access request sent by the user terminal; according to a preset collection requirement, multi-dimensional attributes are extracted from the user request, the user request is converted into an attribute request, and the attribute request is sent to a policy decision point (PDP); the attributes comprise user attributes, environment attributes, operation attributes and object attributes; the PDP performs access permission analysis by using a policy administration point (PAP) based on the attribute request, obtains an access permission analysis result, and feeds back the access permission analysis result to the PEP; the information fed back by the PDP to the PEP comprises permission, rejection, inapplicability and unknown; and the PEP executes the resource access request according to the feedback information received from the PDP. The application is based on a zero-trust technical framework, and fine management of permissions is realized by ABAC, so that the changing business requirements and security challenges can be effectively coped with.

This invention discloses a method and apparatus for querying semantic information of multi-version unstructured data. The method includes: obtaining a query statement; parsing the query statement into an abstract syntax tree; running the node lookup operator at a node lookup node to obtain a specified object attribute calculation node; obtaining version numbers of the unstructured data object and semantic information name at a version extraction node based on the query statement; calculating attribute values ​​of the unstructured data object at the specified object attribute calculation node based on the version number and the unstructured data object; and obtaining a semantic information query result at the semantic information calculation node based on the attribute values ​​of the unstructured data object, the semantic information name, and the version number of the semantic information name. This invention supports querying semantic information of multi-version unstructured data.

The application provides a method and device for placing objects in a game, electronic equipment and a storage medium, comprising: in response to a pre-placing operation on a target object, displaying a ring-shaped grid centered on a center object in a game scene, and displaying the pre-placed target object in the ring-shaped grid; the ring-shaped grid is generated according to the polar coordinate position of the target object in the game scene and the object attribute of the target object, and the polar coordinate position is determined with the center object as the polar point and the ray from the polar point to the preset direction as the polar axis; in response to the first occupied area of the target object in the ring-shaped grid not overlapping with the second occupied area of the placed object, placing the target object in the first occupied area of the ring-shaped grid; by displaying the ring-shaped grid in the game scene and placing the object along the ring-shaped grid, the flexibility of object placement is improved, and the object placement demand in the ring-shaped placement of such a specific game scene is met.

The invention discloses a data processing method, device and equipment and a readable storage medium, historical data sets of multiple business fields can be obtained, and the historical data sets comprise multiple object identifiers and historical attribute data associated with each object identifier; according to the historical attribute data associated with each object identifier, constructing an attribute tag array corresponding to each object identifier, the attribute tag array comprising at least one attribute tag corresponding to the historical attribute data; based on the attribute tag array corresponding to each object tag, obtaining an object tag set associated with each attribute tag; and storing each object identifier in the corresponding object identifier set under each attribute tag into the corresponding compressed bitmap data structure to obtain a target compressed bitmap data structure corresponding to each attribute tag. In this way, the storage resource occupancy rate during object attribute data storage is reduced, and the follow-up multi-dimensional attribute object attribute data query efficiency can be improved.

The invention discloses a digital scene generation method based on digital-real collaborative verification, which belongs to the technical field of digital scenes, and comprises the following steps of: carrying out omnibearing real-time data acquisition on a real physical scene through various sensors, including spatial geometric information, environmental parameters, object attributes and dynamic behavior data, and carrying out preprocessing; and constructing a preliminary model of the digital scene by using the preprocessed data. And performing space-time alignment comparison on the preliminary model and real-time data of a real physical scene, calculating a plurality of verification indexes, performing optimization through reinforcement learning, and dynamically adjusting parameters until a comprehensive difference index is smaller than a threshold value, thereby obtaining a final digital scene model. A block chain verification traceability mechanism is also introduced, and it is ensured that data of each verification link in the digital scene generation process can be queried and traced back. According to the method, efficient and accurate generation and optimization of the digital scene are realized, the consistency and credibility of the digital scene and a real scene are improved, and the method is suitable for the field of scene construction of digital twinning, virtual reality and the like.

Embodiments of the present application provide a data pushing method and device, equipment and medium, the method relates to the field of artificial intelligence, and the method comprises the following steps: receiving business object data uploaded by a first object through a first client; obtaining effective content corresponding to effective object behavior from historical content; taking content attribute values corresponding to the effective content as object attribute values of the first object, determining a weighted business object attribute value of the first object in a business object cycle based on the object attribute values, determining a first attribute tendency of the first object as an attribute tendency corresponding to the weighted business object attribute value; if the weighted business object attribute value meets a reverse attribute recommendation condition corresponding to the business object cycle, obtaining target recommendation content matched with a second attribute tendency, and issuing the target recommendation content to the first client. By adopting the present application, information stream content (i.e. target recommendation content) with an opposite attribute tendency can be pushed to the first object, and the flexibility of information stream content recommendation is improved.