215 results about "Human interaction" patented technology

The present invention introduces a universal AI-powered pet management platform that establishes an entirely new category of technology, transcending conventional pet training systems. This comprehensive system integrates a modular wearable pet device with interchangeable sensors, sophisticated AI processing capabilities, and diverse output modules to create a unified ecosystem for holistic pet care. Unlike traditional training devices focused solely on behavior modification, this platform simultaneously manages multiple domains including real-time health monitoring, environmental safety assessment, emotional well-being analysis, autonomous training, emergency response, and seamless integration with external systems. The platform's universal architecture enables dynamic adaptation across diverse applications from companion animals to service animals, wildlife monitoring, and specialized deployments. By leveraging advanced artificial intelligence models, multimodal communication pathways, and a universal API for third-party integration, the system creates an interconnected technological framework that fundamentally transforms the relationship between pets, technology, and human interaction, rendering isolated pet devices obsolete.

Automated composite agent generation includes generating an agent cooperation graph (ACG) based on machine-human interactions between a user and digital agents pretrained to perform automated tasks. The ACG nodes represent digital agents connected by edges weighted according to cooperation densities. A composite sub-graph (CSG) whose nodes represent cooperating digital agents that cooperatively perform related tasks in response to multiple user requests is generated based on the cooperation densities. A composite agent configured to perform a composite process in response to a composite instruction is generated and includes a composite evaluator and composite actuator. The composite evaluator is generated based on the multiple requests using a language model. The composite actuator is generated using an automated processes compiler that compiles the related automated processes. The composite actuator of the composite agent actuates the composite process, which performs the related automated tasks as a single, integrated process.

Robotic systems and associated methods are described herein. The robotic system may collect measurements from various sensors corresponding to motion of the robotic system, the surrounding environment of the robotic system, or both. The robotic system may generate measurement data based on the collected measurements. Measurements from a particular sensor may be processed in conjunction with different sensors of the robotic system, which may facilitate more accurate or more useful measurement data. The systems and methods of the present disclosure enable the detection, labeling, and locating of features in real time or near real time using the robotic system with little or no reliance on human interaction to detect and map the features. The disclosure provides enhanced accuracy and efficiency as it enhances the functionality and reduces the reliance on human detection of features.

Robotic systems and associated methods are described herein. The robotic system may collect measurements from various sensors corresponding to motion of the robotic system, the surrounding environment of the robotic system, or both. The robotic system may generate measurement data based on the collected measurements. Measurements from a particular sensor may be processed in conjunction with different sensors of the robotic system, which may facilitate more accurate or more useful measurement data. The systems and methods of the present disclosure enable the detection, labeling, and locating of features in real time or near real time using the robotic system with little or no reliance on human interaction to detect and map the features. The disclosure provides enhanced accuracy and efficiency as it enhances the functionality and reduces the reliance on human detection of features.

The invention relates to the technical field of operation error recognition, and discloses an operation error recognition method and system based on human-computer interaction, and the method comprises the steps: collecting man-made interaction behaviors through a front end; carrying out behavior analysis on the interaction behavior, and carrying out primary recognition on errors; transmitting the operation instruction of the interaction behavior which is not judged to be wrong in the primary identification to an execution part, and carrying out secondary error identification based on mechanical constraint in the execution part; analyzing the execution part behaviors to finish the final recognition of errors; after identification is completed, execution partition is performed on each result which is identified as an error, and an execution part rejects an interactive behavior of execution errors, so that operation errors are prevented. Safety and accuracy in complex interaction tasks are greatly improved, and the limitation that in the prior art, operation errors are difficult to detect and process in real time is overcome.

A system includes a hardware processor configured to execute a machine learning (ML) model training pipeline to train an ML model using data relevant to a world of a digital persona to provide a dialogue model, generate, using the dialogue model, first conversational outputs, train the dialogue model, based on the first conversational outputs, to avoid hallucinations and / or undesirable expressions to provide a guardrailed dialogue model, generate, using the guardrailed dialogue model, second conversational outputs, train the guardrailed dialogue model, based on the second conversational outputs and persona data identifying interaction characteristics of the digital persona to provide a persona-specific model, generate, using the persona-specific model, a response to a scripted question, determine a quality score for the response, and further train the persona-specific model or validate the persona-specific model for human interaction, depending upon whether the quality score fails to satisfy or satisfies a quality criterion.

The invention discloses a virtual human interaction method and device, related equipment and a computer program product, and relates to the technical field of artificial intelligence, and the method supports the creation of more than two virtual human roles, can respond to a virtual human role creation request, can create more than two virtual human roles according to role identifiers, and improves the user experience. And respectively generating video streams of the virtual human roles through different virtual human engines. More than two virtual human roles can be controlled at the same time, the corresponding video frames in the video streams of more than two virtual human roles are merged, and the merged target video stream is generated and pushed to the client for playing. According to the method, more than two virtual human roles can be combined into one video stream, the same-screen interaction of more than two virtual humans is realized, the number of the interactive virtual human roles is increased, the diversity of virtual human display is improved, the expressive force of the virtual humans is enriched, the virtual humans can be more vivid in a more complex scene, and the user experience is improved. And personalized requirements of users are met.

Dynamic outputs for inducing interactions with autonomous vehicles is described. In one or more implementations, a system includes an output device configured to produce audible signals in an environment outside an autonomous vehicle, and at least one processor operatively coupled to the output device. The processor is configured to detect one or more characteristics of the environment, dynamically generate, based on the characteristics, an audio output for inducing human interactions with the autonomous vehicle, and cause the output device to transmit the audio output as the audible signals produced in the environment for facilitating a human interaction with the autonomous vehicle.

The invention belongs to the technical field of computer vision, and is mainly developed around character interaction detection (HOI). Aiming at the defects of weak semantic understanding ability, insufficient classifier semantic clue utilization, prominent long-tail distribution problem and the like of an existing HOI detection method in a complex scene, such as difficulty in accurately capturing interactive semantic essence due to shielding and illumination influence, incapability of mining tag semantic association due to random initialization weight, poor detection effect of rare interactive categories and the like, the invention provides a method for detecting the semantic association. The invention provides a character interaction detection method based on multi-modal information fusion. According to the method, a text label semantic enhancement classifier is introduced, multi-modal features are fused into a character interaction detection classification head, a multi-modal instance classifier is realized based on CLIP, and finally, results of the two classifiers are fused. The method specifically comprises the steps of extracting image features through a backbone network, performing set prediction through a Transf ormer encoder-decoder, extracting text embedding features through a multi-modal label semantic embedding module, initializing a linear classification layer weight by using a text embedding vector to perform interactive prediction, performing prediction by using a multi-modal instance semantic classifier, and fusing prediction results through a multi-source fusion module. And finally, a detection result with higher precision is obtained.

Robotic systems and associated methods are described herein. The robotic system may collect measurements from various sensors corresponding to motion of the robotic system, the surrounding environment of the robotic system, or both. The robotic system may generate measurement data based on the collected measurements. Measurements from a particular sensor may be processed in conjunction with different sensors of the robotic system, which may facilitate more accurate or more useful measurement data. The systems and methods of the present disclosure enable the detection, labeling, and locating of features in real time or near real time using the robotic system with little or no reliance on human interaction to detect and map the features. The disclosure provides enhanced accuracy and efficiency as it enhances the functionality and reduces the reliance on human detection of features.

Robotic systems and associated methods are described herein. The robotic system may collect measurements from various sensors corresponding to motion of the robotic system, the surrounding environment of the robotic system, or both. The robotic system may generate measurement data based on the collected measurements. Measurements from a particular sensor may be processed in conjunction with different sensors of the robotic system, which may facilitate more accurate or more useful measurement data. The systems and methods of the present disclosure enable the detection, labeling, and locating of features in real time or near real time using the robotic system with little or no reliance on human interaction to detect and map the features. The disclosure provides enhanced accuracy and efficiency as it enhances the functionality and reduces the reliance on human detection of features.

The invention provides a virtual digital human interaction method and system based on digital twinning, and the method comprises the steps: S1, obtaining an operation instruction of a digital human, carrying out the preprocessing of the obtained operation instruction of the digital human, and obtaining a processed operation instruction of the digital human; s2, triggering a cloud rendering configurator to complete interaction between the digital human and the digital twin system based on the processed digital human operation instruction, and updating the real-time state of the digital twin system; and S3, encoding the updated real-time state of the digital twin system into a pixel stream video, and transmitting the pixel stream video to a client to realize real-time state synchronization of the digital twin system.

Components for a medical infusion fluid handling system, such as an APD system, in which one or more lines (such as solution lines), spikes or other connection ports may be automatically capped and / or de-capped. This feature may provide advantages, such as a reduced likelihood of contamination since no human interaction is required to de-cap and connect the one or more lines, the spikes, or the other connection ports. For example, a fluid handling cassette may include one or more caps that cover a corresponding spike and include a raised and / or recessed feature to assist in removal of the one or more caps from the cassette. A solution line cap may include a hole and a recess, a groove or other feature to engage with a spike cap and enable removal of the spike cap.

Robotic systems and associated methods are described herein. The robotic system may collect measurements from various sensors corresponding to motion of the robotic system, the surrounding environment of the robotic system, or both. The robotic system may generate measurement data based on the collected measurements. Measurements from a particular sensor may be processed in conjunction with different sensors of the robotic system, which may facilitate more accurate or more useful measurement data. The systems and methods of the present disclosure enable the detection, labeling, and locating of features in real time or near real time using the robotic system with little or no reliance on human interaction to detect and map the features. The disclosure provides enhanced accuracy and efficiency as it enhances the functionality and reduces the reliance on human detection of features.

A platform to manage a series of software containers and virtual machines for integrated software development that allows control over data that is exfiltrated or infiltrated through a series of mechanism and network protocols connecting to the container's content, accessible over a network by an authenticated software developer, each software container and virtual machine being associated to a credentials management unit having access to a database of credentials that are not known to the software developer, the credentials management unit being configured to monitor network traffic, detect an authentication process to an external resource in the network traffic, present to the external resource a corresponding credential selected from the database, where data exchange can happen in a purely protocol-based fashion without human intervention via a secure browser to allow control even when human interaction is needed.

Robotic systems and associated methods are described herein. The robotic system may collect measurements from various sensors corresponding to motion of the robotic system, the surrounding environment of the robotic system, or both. The robotic system may generate measurement data based on the collected measurements. Measurements from a particular sensor may be processed in conjunction with different sensors of the robotic system, which may facilitate more accurate or more useful measurement data. The systems and methods of the present disclosure enable the detection, labeling, and locating of features in real time or near real time using the robotic system with little or no reliance on human interaction to detect and map the features. The disclosure provides enhanced accuracy and efficiency as it enhances the functionality and reduces the reliance on human detection of features.

The invention discloses a memory network enhancement method for experimental interaction behavior recognition and application, and belongs to the character interaction behavior recognition technology in video image recognition, a video level graph network is used for extracting global features of each category in a data set, and memory items of the memory network are initialized to serve as supplementary information to reduce the intra-class distance; extracting local features of a single video by using a region-level graph network, capturing detail information of an interaction behavior, and fusing the detail information with global features in a memory network to enrich sample features; and designing memory embedding loss to enlarge the inter-class distance between similar samples, thereby realizing accurate identification of the interaction behavior, and further improving the model robustness. According to the invention, accurate identification of experimental interaction behavior categories in the experimental video is realized, and the overall performance of the model is excellent.

The present invention facilitates communication between sign language users and machines by translating sign language and text using AI models, deep learning computer vision, and word embeddings. Users interact via sign language, captured and processed through deep learning and NLP modules. The system converts sign language videos into text, constructs coherent sentences, and generates contextually appropriate responses using a Retrieve and Generate (RAG) model. Responses are translated back into sign language videos, spelling out words not found in the dictionary. If requested, a human agent can respond. Key features include high-accuracy recognition, context-aware response generation, dynamic vocabulary updates, and optional human interaction. The method ensures efficient processing with LLM, embedding techniques, and deep learning, optimizing translation accuracy and user experience. The system adapts to multiple languages and dialects by training on specific sign languages, making it applicable globally.

The present application relates to an Artificial Intelligence (AI) hand type device that emulates human hand capabilities, delicately to autonomously identify and pick up small and fragile objects, including but not limited to nails, screws, and microchips, with precision. Equipped with advanced sensors, algorithms, and AI, the device accurately locates object positions and angles using computer vision technology and a high-resolution camera. It performs quality inspection analysis, object placement, movement and mechanical operations autonomously, based on AI decisions according to the real time perspective of its surrounding environment.The device's AI-driven robotic arms adapt their grip strength, pressure and temperature for handling tiny delicate items such as eye-glass screws and larger objects such as motorized drill machines. With the ability to learn from domain experts, the present device autonomously executes tasks based on past experiences.Notably, the device allows human interaction through voice commands, user interface, or physical assistance to teach it object identification, precise location detection, quality control readings and mechanical operations to perform a certain skilled task.

A system and method enable pressure testing a specimen with a portable module for deployment to a field site of the specimen, such as a pipeline. The module includes a processor capable of controlling various devices, such as valves and pumps, and recording sensed data from multiple sources, such as pressure / temperature / flow sensors, cameras and weather feed. The processor may receive input operation specifications for execution to perform the testing with limited human interaction through pressurization, hold and depressurization phases of the testing. The module includes safety features that may avoid over pressurization, require operator identifiers or initiate interlock protocols upon the processor detecting an anomaly. The processor then outputs notifications and reports of desired parameters from the test.

Components for a medical infusion fluid handling system, such as an APD system, in which one or more lines (such as solution lines), spikes or other connection ports may be automatically capped and / or de-capped. This feature may provide advantages, such as a reduced likelihood of contamination since no human interaction is required to de-cap and connect the lines, spikes or other connections. For example, a fluid handling cassette may include one or more caps that cover a corresponding spike and include a raised and / or recessed feature to assist in removal of the cap from the cassette. A solution line cap may include a hole and recess, groove or other feature to engage with a spike cap and enable removal of the cap.

The present invention belongs to the field of smart education and virtual reality technology. The present invention discloses a smart teacher system and method for integrated teaching of virtual humans and real teachers. The system includes: an MR access synchronous classroom module, which is used to access the first-person video stream in the MR device into the synchronous classroom for live broadcast; a virtual human interaction module, which is used to interact with the virtual human through voice, and the virtual human interaction module includes a voice control SDK and client voice interaction logic. The present invention can reduce the live broadcast delay, making it convenient for students to observe the teacher's live demonstration through the smart blackboard screen; it can realize the function of users interacting with virtual humans through voice; it can automatically adjust the position of virtual content when the user moves or the environment changes to ensure its synchronization with the real world, provide users with a smooth and efficient remote interaction experience, and realize seamless connection and intelligent assistance of teaching activities.

Robotic systems and associated methods are described herein. The robotic system may collect measurements from various sensors corresponding to motion of the robotic system, the surrounding environment of the robotic system, or both. The robotic system may generate measurement data based on the collected measurements. Measurements from a particular sensor may be processed in conjunction with different sensors of the robotic system, which may facilitate more accurate or more useful measurement data. The systems and methods of the present disclosure enable the detection, labeling, and locating of features in real time or near real time using the robotic system with little or no reliance on human interaction to detect and map the features. The disclosure provides enhanced accuracy and efficiency as it enhances the functionality and reduces the reliance on human detection of features.

The invention relates to the technical field of computer vision, and discloses a figure interaction detection method based on spatial fine-grained context interaction feature fusion, which comprises the following steps of: firstly, performing target detection on an input image to obtain a target detection result set and a person-object pairing feature; performing gridding projection on the image to obtain image global features, and inputting the image global features into a spatial fine-grained feature learning module to obtain spatial fine-grained features; inputting the spatial fine-grained features and the human-object pairing features into a spatial context interaction feature fusion module to obtain spatial context interaction features, and inputting the spatial context interaction features and the spatial fine-grained features into a visual encoder to obtain enhanced human-object pairing features; obtaining an interaction category score according to the feature and a text embedding feature of a character interaction category; and iteratively optimizing the character interaction detection model until convergence. According to the method, the description of local details of a visual space is enhanced, the context relationship between a person and an object is enhanced, and the accuracy of person interaction detection is improved.

The invention discloses an intelligent doctor seeing method, system and device for AI virtual person accompanying diagnosis and a medium. The method comprises the steps that an intelligent hospital guide demand instruction is received, a virtual consulting room is constructed, and an AI virtual person is generated; after it is judged that the user enters the virtual consulting room, user input information is obtained in real time; determining a user object type and a user hospital guide application scene according to the user input information; the abnormal emotion of the user is recognized, and a preset emotion pacifying strategy is matched, so that the AI virtual human interacts with the user according to the matched emotion pacifying strategy; according to the real-time interaction information of the user and the historical intelligent hospital guide information of the user, matching a preset deep learning model in combination with a hospital guide application scene of the user, and outputting AI virtual human interaction reaction information and a preliminary diagnosis result; and determining a registration department according to the preliminary diagnosis result, generating registration appointment recommendation and completing registration appointment. According to the invention, the user experience in the intelligent treatment process of AI virtual human accompanying diagnosis can be improved.

The invention provides a virtual human interaction action generation method for unmarked augmented reality equipment, and the method comprises the steps: obtaining a head posture, a hand posture and a first-person visual angle image sequence of a user through a built-in sensor of an augmented reality helmet, and jointly reconstructing a whole-body posture of the user on the basis; generating an interaction action of a virtual human according to the reconstructed user posture; according to the method, the interactive action of the virtual human can be more natural and real and is kept consistent with the actual behavior of the user; in the field of man-machine interaction, the prior art depends on user control and needs additional hardware equipment and manual participation to maintain an interaction process; or depending on predefined conditions, generalization and practicability are poor; compared with the prior art, additional manual intervention and predefined design are not needed, interaction with the natural response virtual human can be achieved only through the lightweight augmented reality helmet, hardware dependence and labor cost are reduced, automation and usability of the method are promoted, and interaction convenience and immersion are improved.

The invention belongs to the technical field of smart education and virtual reality, and discloses a smart teacher system and method for fusion teaching of virtual people and real teachers, and the system comprises an MR access synchronous classroom module which is used for accessing a video stream of a first visual angle in MR equipment into a synchronous classroom for live broadcast; the virtual human interaction module is used for interacting with a virtual human through voice, and the virtual human interaction module comprises a voice control SDK (Software Development Kit) and client voice interaction logic; according to the invention, the live broadcast time delay is relatively small, and students can conveniently observe the live broadcast demonstration of teachers through the smart blackboard screen; the function that the user interacts with the virtual human through voice can be realized; according to the technical scheme, when the user moves or the environment changes, the position of the virtual content can be automatically adjusted, synchronism of the virtual content and the real world is ensured, smooth and efficient remote interaction experience can be provided for the user, and seamless connection and intelligent assistance of teaching activities are achieved.

The invention relates to the technical field of advertisement design, and discloses an AI virtual human interaction method and system, and the method comprises the steps: receiving original demand data inputted by a user, generating a brand feature vector, and generating first virtual human data, first line data and first product model data based on the brand feature vector; and performing cooperative adjustment based on the ith interaction instruction to generate a virtual human image. By implementing the method and the system, the functions of full-process intelligent design, data processing, collaborative generation, interactive response, scene adaptation and the like are improved, an efficient and complete virtual human image customization scheme is formed, and the problem that multi-source heterogeneous data are difficult to comprehensively analyze when user requirements are processed by a traditional method is solved. The problems of large deviation between generated content and user expectation and poor scene adaptability caused by lack of a collaborative adjustment mechanism of a virtual human image, lines and a product model, low processing efficiency of a modification instruction and incapability of ensuring cross-modal consistency in the prior art are solved.

The invention discloses a human interaction behavior identification method and device for emergency and rescue, and the method comprises the steps: obtaining video data captured by an unmanned plane, obtaining a plurality of video frames, extracting key frames in the plurality of video frames, and determining a plurality of sliding windows of the video frames; carrying out human behavior recognition on the key frame by adopting a convolutional neural network to generate a feature vector, constructing a probability vector to obtain a feature set, carrying out prediction analysis on the feature set by adopting a support vector machine algorithm to obtain a corresponding action label, and recognizing human behaviors and marking corresponding actions according to the action label. The efficiency of rescue and emergency actions is improved by using the video clips of the unmanned aerial vehicle, the man-machine interaction recognition in a disaster scene is improved, human-free detection, tracking or multiple image instances are improved, the efficiency in the aspects of rescue and emergency management response accuracy, processing time and calculation complexity is improved, and the rescue and emergency management efficiency is improved by analyzing the first image and the last image. The image quality is improved, and disaster management and rescue actions are accelerated.

Dynamic outputs for inducing interactions with autonomous vehicles is described. In one or more implementations, a system includes an output device configured to produce audible signals in an environment outside an autonomous vehicle, and at least one processor operatively coupled to the output device. The processor is configured to detect one or more characteristics of the environment, dynamically generate, based on the characteristics, an audio output for inducing human interactions with the autonomous vehicle, and cause the output device to transmit the audio output as the audible signals produced in the environment for facilitating a human interaction with the autonomous vehicle.