10 results about "Dual modality" patented technology

The application discloses an AIGC interaction experience quality evaluation method and system based on eye movement and brain electrical dual modalities, relates to the technical field of man-machine interaction evaluation and cognitive neuroscience, and comprises the following steps: collecting brain electrical and eye movement signals of a user performing an AIGC interaction task in an entire process, and synchronously recording all key interaction events; pre-processing the signals, dividing the signals by taking the key interaction events as anchor points, and constructing event-related time sequence data segments; for each segment, carrying out dual-modality time sequence feature extraction by using an eye movement flow encoder and a brain electrical flow encoder, fusing and splicing the extracted brain electrical and eye movement features, and obtaining an interaction fluency hidden vector; based on the brain electrical and eye movement features and the interaction fluency hidden vector, combining preset indexes, and respectively calculating original scores of fluency, satisfaction and usability; constructing a scene vector according to a user interaction scene, generating weights through a dynamic adaptive weight network, dynamically weighting each original score, and obtaining a comprehensive score of the interaction experience quality.

The invention provides a bimodal integrated device for particle radiotherapy. The bimodal integrated device comprises a control system, a sitting and standing posture system, a bimodal imaging system and an image processing system, the sitting and standing posture system is used for bearing and driving a to-be-detected object to rotate; the bimodal imaging system comprises at least two scanning units composed of ray source assemblies and detection assemblies to obtain image data, the image data comprises first data and second data, the first data comprises two-dimensional image data of a to-be-detected object, and the second data comprises three-dimensional image data of the to-be-detected object rotating to a preset angle; the bimodal imaging system outputs acquired image data to the image processing system; and the image processing system is used for generating medical images for monitoring and registration according to the image data, and has the advantages of being suitable for image guidance and real-time monitoring of sitting and standing radiotherapy while giving consideration to the three-dimensional anatomical precision and the two-dimensional instantaneous speed.

The present application relates to the technical field of mild cognitive impairment early screening, and discloses a dual-mode signal processing method for intelligent screening of mild cognitive impairment, which comprises the following steps: acquiring electroencephalogram signals and eye movement signals of a target object under the driving of a cognitive task; registering the electroencephalogram signals and the eye movement signals according to the cognitive task to obtain electroencephalogram signals and eye movement signals with the same time sequence length; and minimizing statistical differences of the registered electroencephalogram signals and eye movement signals in each corresponding time window and maximizing time sequence correlation of the registered electroencephalogram signals and eye movement signals in each corresponding time window to obtain target electroencephalogram signals and target eye movement signals with consistent distribution attributes; wherein the statistical differences include mean difference, variance / standard deviation difference and distribution shape difference. The dual-mode signal processing method for intelligent screening of mild cognitive impairment has a low data acquisition cost, and the obtained signals have high accuracy in representing cognitive characteristics.

Switching a position of a dual modality coupling module to reduce vibrations in a machine includes simulating the propagation of vibrations through the machine using a digital twin model. The simulation is used in determining vibration levels at points of interest in the machine and a recommendation is generated to use the dual modality coupling module based on the vibration levels and the defined vibration threshold. If the vibration levels fail to satisfy the vibration threshold, the dual modality coupling module is positioned at the mechanical coupling position. Upon determining that the vibration levels satisfy the defined vibration threshold, a signal is generated to regulate a current flow to magnetic coils of the dual modality coupling module. The current flow induces temporary magnetism in the magnetic coils. As a result, a mechanical coupling position is switched to a magnetic coupling position.

A dual modality relationship network for audiovisual event localization can be provided. A video feed for audiovisual event localization can be received. Based on a combination of extracted audio features and video features of the video feed, information features and regions in the video feed can be determined by running a first neural network. Based on the information features and regions in the video feed determined by the first neural network, relationship-aware video features can be determined by running a second neural network. Based on the information features and regions in the video feed, relationship-aware audio features can be determined by running a third neural network. A dual modality representation can be obtained based on the relationship-aware video features and the relationship-aware audio features by running a fourth neural network. The dual modality representation can be input to a classifier to identify an audiovisual event in the video feed.

The disclosed techniques provide a dual modality for private chat sessions for populating a private thread during a meeting with messages that are communicated in a private thread prior to a meeting. During the lifecycle of a meeting (before, during, and after a meeting), the system allows a participant to initiate, via the chat modality, a private chat among a subset of meeting attendees associated with a meeting. A first private thread is created prior to a meeting, and those message are communicated in the first thread before a start time of the meeting. Then, during the meeting, the system creates a second private thread while in the meeting modality, e.g., showing a meeting UI that uses a meeting object. If the users of the second private thread are the same as the first private thread, the messages of the first private thread are populated in the second private thread

The invention discloses a spokesman tracking method based on audiovisual bimodal and related equipment, and relates to the technical field of data processing, and the method comprises the steps: obtaining audio information and video information in response to a spokesman tracking instruction, carrying out voiceprint comparison based on a historical database and the audio information, and obtaining a spokesman tracking result; the method comprises the following steps: acquiring a historical database, determining whether a voice ID corresponding to a spokesman exists in the historical database, the historical database comprising a plurality of voice IDs and face IDs corresponding to the spokesman, and determining the spokesman based on the voice ID corresponding to the spokesman if the voice ID corresponding to the spokesman exists in the historical database; and if the video information does not exist, performing lip movement identification based on the video information so as to determine a spokesman. According to the method, multi-dimensional information such as sound source comparison and lip movement detection is comprehensively utilized, and rapid confirmation and stable tracking of the identity of the spokesman in a complex environment are realized through cross-modal feature fusion.

This invention discloses a dual-modal support device, a control method for the dual-modal support device, and a medical trolley. The dual-modal support device includes a moving mechanism, a joint module, a lifting mechanism, a dual-modal release module, and a support mechanism. The joint module is connected to the moving mechanism. One end of the lifting mechanism is connected to the joint module, and the joint module can drive the lifting mechanism to rotate relative to the moving mechanism. One end of the dual-modal release module is hinged to the moving mechanism. One end of the support mechanism is connected to the dual-modal release module, and the other end is hinged to the end of the lifting mechanism opposite to the joint module. The support mechanism can adaptively extend and retract with the rotation of the lifting mechanism relative to the moving mechanism, and supports and locks the lifting mechanism when it rotates to a preset angle. The dual-modal release module includes a mechanical unlocking component and an electric unlocking component. The electric unlocking component can apply force to the support mechanism to unlock it; a manual force is applied to the mechanical unlocking component to unlock the support mechanism.

Switching a position of a dual modality coupling module to reduce vibrations in a machine includes simulating the propagation of vibrations through the machine using a digital twin model. The simulation is used in determining vibration levels at points of interest in the machine and a recommendation is generated to use the dual modality coupling module based on the vibration levels and the defined vibration threshold. If the vibration levels fail to satisfy the vibration threshold, the dual modality coupling module is positioned at the mechanical coupling position. Upon determining that the vibration levels satisfy the defined vibration threshold, a signal is generated to regulate a current flow to magnetic coils of the dual modality coupling module. The current flow induces temporary magnetism in the magnetic coils. As a result, a mechanical coupling position is switched to a magnetic coupling position.

This invention relates to the field of intelligent systems technology and discloses a dual-modal fusion system for assessing and screening the risk of diabetic retinopathy. The system includes: a user terminal for acquiring fundus images and non-image risk feature data of the user; and a cloud server terminal including: an image analysis module for detecting lesions in the fundus images based on a deep learning model and outputting image assessment results; a structured data analysis module for quantitatively scoring the non-image risk feature data based on a preset risk assessment model and outputting questionnaire assessment results; and a comprehensive assessment engine module for fusing the image assessment results and the questionnaire assessment results using a decision-level weighted fusion strategy to generate a comprehensive risk assessment result. This invention can take into account both image morphological information and multi-dimensional clinical risk features, achieving comprehensive and accurate risk assessment with good interpretability and robustness.