18results about "Multimedia data retrieval" patented technology

The present disclosure relates to a mobile terminal having an artificial intelligence unit, and a control method therefor. A mobile terminal according to the present disclosure includes: GPS configured to receive location information of the mobile terminal, a wireless communication unit configured to perform wireless communication with an external device, and a controller configured to store communication information when a communication event with the external device is generated. The controller matches the location information received from the GPS with the communication information and stores the matched information when the communication event is generated, and outputs at least one information corresponding to a search query for retrieving communication information from pre-stored communication information by using location information matched to the pre-stored communication information and the search query, when the search query is input by a user.

Methods and systems for suggesting image modifications using one or more machine learning models are disclosed. The methods and systems select, by an interactive application, individual content items from a plurality of previously captured content items that match one or more criteria corresponding to sharable content, and generate prompts that include the individual content items and a request for a plurality of suggested modifications of the individual content items. The methods and systems process cues through a large language model (LLM) to generate a plurality of suggested modifications to individual content items, and generate modified individual content items corresponding to individual suggested modifications of the plurality of suggested modifications.

A method and system (100) for managing video sharing in a collaboration session is disclosed. A sharing function (112) obtains a synchronization timestamp of an audio frame of the collaboration session (B122), which is used to synchronize a shared video stream with the collaboration session. The sharing function (112) determines whether the shared video stream requires more, fewer, or the same number of video frames as the video file, based on the video frame rate difference between a desired frame rate and the source frame rate of the video file. The sharing function (112) iteratively retrieves video frames from the video file stored in a storage function (140) (B140). The sharing function (112) provides the shared video stream by sending one or more generated video frames to a video communication service (110) (B160). The video communication service (110) provides the shared video stream to two client devices (121) participating in the collaboration session (B174).

Disclosed are a system, method and software to associate attributes with digital media assets. Digital media contains specific assets, such as images, that can be replaced with other assets. The system, method and software permit the association of attributes with specific assets. The association of attributes and assets enables the provision of content that is enhanced and more impacting for a user.

An information processing device (10) includes: an extraction unit (11) that extracts text information representing a sound from an image published in a comic for each comic processing unit; a comparison unit (12) that compares the extracted text information for each comic processing unit with text information representing a sound in an anime scene prepared in advance, and computes a similarity; and a correspondence derivation unit (13) that obtains a correspondence relationship between the anime scene and the comic processing unit based on the computed similarity.

We provide the system. [Solution] A speech recognition means that converts speech data received from a speech input device into text data, Image recognition means that analyzes image data received from an image acquisition device and assigns labels based on specific events, A natural language generation means that automatically generates text based on converted text data and labeled image data, A database management system that organizes and saves generated text in a diary format, A system that includes this.

A method and system (100) for managing video sharing in a collaboration session is disclosed. A central server (130) obtains synchronization timestamps of audio frames associated with the collaboration session, which are used to synchronize the shared video stream with the collaboration session (B122). The central server (130) then determines whether more, fewer, or the same number of video frames are needed based on the video frame rate difference of the video files (B124). The central server (130) sets up a buffer for providing the shared video stream to the collaboration session (B130). This buffer is capable of buffering a limited number of video frames corresponding to time intervals of less than 1 second at most. The central server (130) iteratively obtains the video frames of the video file frame by frame (B140). For each video frame, the central server (130) generates (B150) zero, one, or more video frames based on the determination, and sets a timestamp for each of the generated one or more video frames. The generated one or more frames have a desired resolution, and each timestamp corresponds to a desired frame rate. Next, the central server (130) provides the shared video stream by inserting one or more generated video frames into a buffer (B160).

Methods, systems, and storage media for providing feedback regarding augmented reality device content are disclosed. Exemplary implementations may: detect, via a sensor of an augmented reality device having an outwardly facing camera, that a user of the augmented reality device appears within content presented in a view area thereof; responsive to detecting that the user of the augmented reality device appears within the content presented in the view area of the augmented reality device, determine an action being performed by the user; and responsive to determining the action being performed by the user, supplement the content presented in the view area of the augmented reality device.

The present invention allows a CEC system to automatedly track the use, storage, access, and modification of sensitive information / data in the system through desktop monitoring. Further, through desktop, video, and audio monitoring of CSRs the system can automatedly determine the improper use, access, storage, and modification of sensitive information by implementing sensitive data use rules that allow a system to be specialized for the user. Finally, the system can automatedly determine and implement violation actions for the improper use, storage, access, and manipulation of sensitive information. This provides an intelligent system capable of locating all sensitive data in the system and regulating the use, access, and storage of sensitive data in the system.

A computer implementation for managing files is disclosed. This method includes rendering at least a portion of a file as an image by opening the file (E110) using a rendering function (610), and obtaining one or more event messages (E120). For each of the one or more event messages, the method includes: obtaining an image corresponding to each event message from the rendering function (610) (E130), where the image represents the rendered portion of the file and is associated with a sequence instruction, generating metadata based on the event message and sequence instruction (E140), providing the metadata (E150), and providing a video containing a set of images (E160), where the set of images contains an image corresponding to each event message, and the video is a representation of the file.

According to embodiments of the present disclosure, a method, apparatus, device, a medium and a program product for media content generation are provided. The method includes: determining, in response to receiving a media content generation request of a user, at least one keyword corresponding to at least one factor of a plurality of factors related to media content generation based on the media content generation request; determining a material set matching the at least one keyword from a media content material library; generating a description text for media content to be generated based on the material set and the at least one keyword corresponding to the at least one factor; selecting a plurality of media content materials matching a story description from the material set; and generating media content based on the plurality of media content materials.

Methods and systems are disclosed for suggesting scenes for an image using one or more machine learning models based on LLM output. The methods and systems generate a first cue including a person's demographics and a date via a user's device, and process the first cue using a large language model (LLM) to generate multiple ideas related to that person on that date, each idea including a corresponding description and atmosphere. The methods and systems generate a second cue including an idea selected from the multiple cuees and a request for multiple scenes related to the selected idea, process the second cue using an LLM to generate multiple scenes related to the selected idea, and present individual content items corresponding to individual scenes among the multiple scenes.