10 results about "Action Code" patented technology

The present disclosure relates to systems, methods, and non-transitory computer-readable media that perform text-to-image editing using executable code generated from natural language text input. For instance, in one or more embodiments, the disclosed systems receive, from a client device, a digital image and natural language text input providing instructions for modifying the digital image. The disclosed systems also generate, using a large language model, executable action code for modifying the digital image in accordance with the instructions of the natural language text input, the executable action code being compatible with an editing application. The disclosed systems further modify the digital image by executing the executable action code via the editing application and provide the modified digital image for display via a graphical user interface of the client device.

The invention relates to the technical field of multi-modal interaction, in particular to a multi-modal interaction control method and system for a sweeping robot, and the method comprises the following steps: collecting voice and gesture input, extracting voice trigger information and action features, generating a modal frequency mark and an instruction list, matching a task path, and generating a modal identification table. The method comprises the following steps: extracting and comparing action sequences to recognize a master control mode, analyzing continuity to generate a response mark, and screening consistent input to generate an interaction instruction result. In the method, through extracting time, quantity and confidence characteristics of voice and gesture input, establishing frequency and trigger marks of multi-mode input, realizing accurate recognition of an input state; a task path and semantic content are combined to carry out space and semantic matching, the task relevance judgment capability is improved, based on action code comparison, dominant modes are identified, conflict instructions are eliminated, the master control status of each mode in continuous tasks is dynamically balanced, the control response accuracy is improved, and an interaction mechanism with intelligent judgment capability is constructed.

The invention discloses a self-calibration IMU intention recognition method and system, and belongs to the technical field of man-machine interaction. The method aims at solving the problems that in the prior art, due to drifting of a sensor, the long-term stability is poor, and due to the fact that an action process and a static posture cannot be distinguished, an instruction is mistakenly triggered and repeated. In order to solve the problem, the method provided by the invention comprises the following steps: acquiring a motion data stream of a sensor, and segmenting the motion data stream into analysis windows; independently determining an action code for representing a dynamic process in the window based on the dynamic characteristics in the analysis window; independently determining a state code for representing the static posture of the tail end of the window based on the static characteristics in the analysis window; and generating a control instruction based on the logic association verification between the action code and the state code. In addition, drift is suppressed through a zero-speed updating mechanism, and drift is actively compensated by resetting a yaw reference when the upright posture is detected.

The invention discloses an intelligent routing jump recommendation method based on reinforcement learning, and relates to the technical field of computer networks. Comprising the steps that 1, multi-dimensional data are collected, the multi-dimensional data comprise user role information, page types and user feedback data, 2, a state space S is defined, the collected data are subjected to structured coding, the state of each system is represented by a three-dimensional vector, all dimensions correspond to user role codes, page type codes and user feedback state codes respectively, and the user role codes, the page type codes and the user feedback state codes are stored in the state space S; 3, creating a Q table, initializing a Q value table according to the size of the state space S and the size of the action space A, the dimension of the Q table being S * A, and initializing each element Q (s, a) into a fixed value Q0, 4, carrying out action coding, carrying out unique coding on a jump target of each page, and establishing a mapping relation table of action codes and actual page jump logic, 5, optimizing the jump action based on a recommendation strategy of Q-Learning; and 6, generating and outputting a recommendation result, and receiving feedback.

This invention discloses a method and apparatus for cross-project security remediation detection and migration in the blockchain ecosystem, relating to the field of blockchain technology. The method includes: constructing a code submission knowledge graph with code submission as the central node and edge relationships such as submitted modified functions, submitted related topics, submitted related merge requests, submission belonging to a project, project reuse of a project, and merge request discussion topics; determining the editing action based on the syntax tree differences of the affected functions to which the code submission belongs, and determining the action code sequence by backfilling the original source code; outputting the remediation probability of a code submission belonging to a security remediation submission using the submission message and line-level differences, the code submission evidence subgraph of the code submission knowledge graph, and the action code sequence; and constructing cross-project migration rules and performing project migration detection when the remediation probability is greater than a probability threshold to determine the cross-project vulnerability status. Based on the above scheme, the identification and migration performance of cross-project security remediation in the blockchain ecosystem is improved.

Securing a file in a computer network includes processing a file system request that identifies a user entity, a file entity, and a file-directed action. Responding to the request with an indication of permission to reveal a portion of the file system, an indication of permission to reveal at least a portion of the file, and an indication of permission to perform the file-directed action. The response being based on access and use constraints of the file entity, as well as a portion of user entity-context and action code entity-context.

The application discloses a continuous action semantic coding and translation method and system for a humanoid robot, and the method comprises the following steps: establishing a unified reference time axis, and setting a sampling and coding element action data structure and a decoding element action data structure; sampling data of each actuator according to the sampling and coding element action data structure; coding the sampling points according to action semantics, and outputting the coding and storing the coding in an action library after the coding is completed; taking out the action coding stream from the action library when the robot moves, and decoding the action coding stream; calculating the continuous action coding stream by using an action smoothing algorithm after the decoding, and filling each action point of the continuous action coding stream according to the decoding element action data structure; and outputting the decoded action coding stream and executing the decoded action coding stream according to the unified time axis. The joints of the humanoid robot can independently move at the same time point, and the linkage action of the humanoid robot can be realized on the whole.

The present application provides a method, device and system for fine-grained collection of equipment action data, which relates to the field of digital information transmission technology. The method includes: when an action start signal or an action end signal of a production line equipment is monitored, an action recording event is triggered, the action recording event is encoded through a preset encoding protocol to obtain an action code, and the coding time is obtained according to the triggering time of the action recording event; the action recording data of the action recording event is obtained according to the action code and the coding time; the action recording data is transmitted to a data storage terminal; the data storage terminal is used to parse the action recording data to obtain the action code and the coding time, and the equipment action data is obtained according to the action code and the coding time. The method of the present application improves the fine-grainedness of the equipment action data and ensures the association between the equipment action data and the action of the production line equipment.

The invention relates to an algorithm visualization method based on a large model. The method comprises the steps of converting to-be-visualized target data through a large model to obtain an original code, and analyzing the original code to obtain an analysis result; identifying the analysis result to obtain at least one insertion point in the original code and the position and the type of the insertion point; matching a target operation code template from a preset code library according to the type of each insertion point in sequence; sequentially generating calling codes according to the positions of the insertion points and the corresponding target operation code templates, inserting the calling codes into the positions of the insertion points, converting the calling codes into operation objects, and arranging the operation objects according to an execution sequence to generate an operation queue; and visualizing the target data based on the operation queue. According to the scheme, semantic analysis, insertion point positioning and insertion code generation are automatically performed in the visualization process by using the large model, so that the insertion accuracy and consistency of the insertion codes in the visualization operation are ensured, and the applicability of the visualization technology is enhanced.

The application discloses a method for training an action generation model, an action generation method, an action generation device and an action generation equipment, and relates to machine learning of artificial intelligence. The method comprises the following steps: calling an action encoder to perform action feature extraction on an action state of a virtual character to obtain an action code; calling a style encoder to perform style feature extraction on the action state of the virtual character to obtain a style code; calling a decoder to decode the action code and the style code to obtain a predicted action of the virtual character; calculating a style loss according to the style codes of different action states; calculating an action loss according to the action codes of different action states; calculating a reconstruction loss according to a real action corresponding to an action state and the predicted action; and training the action generation model according to the action loss, the style loss and the reconstruction loss. The method can distinguish the action styles of different virtual characters and improve the action quality of the generated virtual characters.