100 results about "Video Media" patented technology

The invention discloses an AI automatic editing method and related equipment, and the method comprises the steps: obtaining a video prototype template selected by a user, prompt information inputted by the user, and original materials uploaded by the user through a user interaction interface, and calling a language large model to generate a structured video script; performing association degree matching on the structured script and the original material through a target detection model, and screening the original material according to a matching result and a quality scoring model to obtain an original video clip; performing background music matching according to the video script to obtain background music, generating voice dubbing and a flower-character chartlet according to the video script, generating video subtitles according to the voice dubbing, performing time axis alignment on each video media element, and synthesizing into a finished video. According to the embodiment of the invention, full-process automatic processing from script generation to video synthesis can be realized, and a high-quality finished video meeting user requirements can be efficiently output. The method can be widely applied to the technical field of automatic editing.

An example apparatus for decoding media data includes: a memory configured to store video data; and a processing system comprising one or more processors implemented in a circuit, the processing system configured to instantiate a first number of video decoder instances to be executed by the processing system; determining an attribute of the plurality of video media streams, the attribute indicating that each of the plurality of video media streams is available for stream selection; selecting the second number of input video media streams from the plurality of video media streams according to the determined attributes of the second number of input video media streams; executing the video decoder instance to decode the second number of input video media streams to form a second number of decoded video media streams; and outputting data of the second number of decoded video media streams.

The invention discloses an AI engine docking decision-making method based on a 5G new call, and the method comprises the steps: carrying out the call service, and analyzing the call information of a user; the AI engine provides an intelligent interaction analysis capability; the media docking module is used for receiving a function instruction of calling service logic, providing basic media capability and making and displaying media content according to user operation; the BM module is used for storing user services and setting data; wherein the call service synchronizes a call instruction, audio and video media stream information and user behavior data to the media docking module, and receives a media instruction. According to the method, real-time multi-mode interactive perception and dynamic AI engine docking can be realized, so that the bottleneck in the prior art is broken through.

The invention provides a scene recognition and value-added service adaptation system and method based on a 5G new call, and aims to improve the intelligence and automation level of the 5G call. The system comprises a service control processing module, a service function intelligent selection module, a voice media processing module, a video media processing module, an AI intelligent identification module, a user interface module and an RTP processing module. User behaviors and call content are analyzed through the AI technology, and the system can automatically recognize call scenes and intelligently select corresponding value-added services such as real-time translation, voice transcription and virtual background. The user interface module optimizes scene recognition and automatic switching functions, provides a value-added service control panel and personalized setting, supports voice instruction operation, and realizes efficient interaction between a user and the system. The system can seamlessly integrate the value-added service into the call media stream, dynamically adjust the service to adapt to the change of the call scene, and continuously optimize the call experience of the user.

A non-transitory computer readable medium stores a video media bitstream encoded by an encoding method. The encoding method includes determining a filter shape of a Cross-Component Sample Offset (CCSO) filter, and determining an offset value associated with the filter shape of the CCSO filter. In the encoding method, the CCSO filter is applied on a sample based on the offset value associated with the filter shape. Filter shape index information is encoded in the video media bitstream. The filter shape index information indicates which of a plurality of filter shapes to apply to the sample. The plurality of filter shapes includes a plurality of 3-tap filter shapes.

A media application obtains a video that includes an audio portion. The media application separates the audio portion into a plurality of channels, where each channel corresponds to a particular audio source. An on-screen classifier model obtains an indication of whether the particular audio source for each channel is depicted in the video. An audio-type classifier model determines, an auditory object classification for each channel. The media application determines a respective gain for each channel based on the indication of whether the particular audio source for the channel is depicted in the video and the auditory object classification for the channel. The media application modifies each channel by applying the respective gain. The media application mixes the modified channels with the audio portion to generate a combined audio.

A method for segmentation prediction implemented by an encoding / decoding device. The method includes obtaining a probability of the segmentation type to be implemented when segmenting an image; and skipping one or more segmentation types based on the obtained probabilities when segmenting encoding / decoding blocks of the image during a conversion between a video media file and a bitstream. Corresponding apparatus and a non-transitory computer-readable medium are also provided.

The invention relates to a media file generation method and device, equipment and a medium, and the method comprises the steps: obtaining a first image and a first text, and generating a plurality of second images based on the first image and the first text; the first text is a picture description text of the media file to be generated; generating a media file based on at least one first main body object in the plurality of second images and the plurality of second images; the media file is an image or a video, and the media file presents a spliced picture obtained by splicing at least part of pictures corresponding to a plurality of second images; in the spliced picture presented by the media file, the object area of the same first main body object covers partial pictures corresponding to the at least two second images respectively. According to the method, the media file with a multi-picture splicing effect and a main body object cross-picture presentation effect can be obtained, the content creation capability of multimedia editing software can be well improved, and diversified media file creation requirements of users can be met.

A mechanism for processing video data is disclosed. A parent block is partitioned with an Extended Ternary-Tree (ETT) partition to create three sub-blocks. At least one of the sub-blocks includes a side with a measurement that is not a power of two. A conversion is performed between a visual media data and a bitstream based on the sub-blocks.

A video recommendation method fusing social information comprises the steps that firstly, a user-video interaction graph and a social graph are constructed through video media platform data, different combination strategies of a diffusion model and graph convolution operation are applied in a denoising priority path and a structure priority path respectively, and collaborative perception and fusion of user embedding and structure information are achieved; secondly, aligning user embedding output by the double-track denoising path by adopting comparative learning, and enhancing robustness; secondly, performing deep coding on the user-video interaction graph and the social graph by applying a double-graph neural network, comprehensively capturing user preferences and embedding videos; and finally, for noise possibly introduced by multi-module fusion, deep denoising is executed to purify user embedding again. According to the method, guidance of structural information is introduced in the diffusion denoising process, and more robust user embedding is obtained by using a comparative learning strategy, so that a structural perception high-fidelity social denoising task is completed, and personalized video recommendation is realized on a video media platform.

Methods and systems are herein provided for judder adjustment. In one example, a system comprises a display device configured to display scenes of a video media and a user interface; and a computing device configured to determine a selected judder angle for a reference display scenario; adjust the selected judder angle for a target display scenario based on the one or more parameters of the scene for two or more regions, wherein the adjusted judder angle corresponds to a judder level of the target display scenario according to the one or more parameters; interpolate one or more frames into one or more of the plurality of input frames based on the adjusted judder angle; and save the scene with the interpolated one or more frames for the target display scenario to memory.

A non-transitory computer readable medium stores a video media bitstream encoded by an encoding method. The encoding method includes determining a plurality of intra prediction modes for a zero reference line in a plurality of reference lines. The zero reference line is closest to a current block among the plurality of reference lines. The current block is predicted with intra prediction. A length of a most probable mode (MPM) list is determined based on a current reference line. The encoding method includes determining at least one MPM included in an MPM list for a non-zero reference line in the plurality of reference lines.

Embodiments of the present application relate to the technical field of communication, and provide a video greeting playing method, system, server and storage medium. The video greeting playing method includes: in response to receiving a call transfer request indicating that a first terminal fails to call a second terminal, acquiring a type of a media channel currently accessed by the first terminal; in response to that the type of the media channel currently accessed by the first terminal is audio, performing video media negotiation with the first terminal; and in response to that the video media negotiation with the first terminal succeeds, sending a video greeting pre-recorded by the second terminal to the first terminal for the first terminal to play the video greeting.

Aspects of the disclosure provide methods, apparatuses, and non-transitory computer-readable storage mediums for video encoding / decoding. A non-transitory computer readable medium stores a video media bitstream encoded by an encoding method that includes selecting a secondary transform core for coding a current block. The secondary transform core has a size of M×N. The encoding method includes applying a forward primary transform to a transform unit of the current block to generate a primary transform coefficient block having a size of W×H. One of H or W is less than M and N. The secondary transform core is applied to the primary transform coefficient block by applying a sub-section of the secondary transform core to the primary transform coefficient block. A secondary transform coefficient block is generated. The encoding method includes encoding, in the video media bitstream, the current block based on an intra prediction mode and the secondary transform coefficient block.

In one embodiment, a method includes receiving an advertisement exchange requesting a rewarded video advertisement from an external server, tokenizing the advertisement exchange based on coding schemes, unwrapping wrappers associated with the advertisement exchange to fetch an original video media file, generating a re-encoded video media file based on re-encoding the original video media file, wherein generating the re-encoded video media file comprises integrating visual elements into the original video media file, wherein the visual elements are configured for enabling the rewarded video advertisement, generating the rewarded video advertisement based on replacing the original video media file with the re-encoded video media file, and sending the rewarded video advertisement to the external server.

This invention discloses a method, system, electronic device, and storage medium for displaying translated subtitles during phone calls. The method includes: triggering the network side to complete audio and video media anchoring and establishing an ADC channel between a first calling terminal and the network side; responding to a translation start command from the first calling terminal, initiating a media stream copying request to the data channel signaling function network element via an intelligent translation application server, causing the media function network element to transmit the uplink audio stream from the second calling terminal to the media processing function network element; translating the uplink audio stream to obtain a subtitle stream via the media processing function network element and returning the subtitle stream to the media function network element; and combining the subtitle stream with the downlink video stream to obtain a subtitle video stream for display. This invention improves call efficiency and the accuracy of information acquisition and can be widely applied in the field of communication technology.

The present invention provides a method, device, and storage medium for switching an audio call to a video call, relating to the field of communication technology, and for improving the smoothness of the switching process between an audio call and a video call. The method comprises: receiving first request information from a user terminal requesting that a call between the user terminal and a service terminal be switched from an audio call to a video call; in response to receiving the first request information, sending a second request information to a video service system requesting the video service system to reserve system resources required for the video call between the user terminal and the service terminal; receiving second response information from the video service system, the second response information being used to determine that the system resources have been reserved; and in response to receiving the second response information, sending a third request information to the service terminal instructing the service terminal to use the system resources to create a video media channel, so that the service terminal and the user terminal can conduct a video call via the video media channel.

A method implemented by a video coding device includes applying a neural network (NN) filter to unfiltered samples of a video unit to generate filtered samples. The NN filter is based on a NN filter model configured to obtain an attention based on a coding parameter input. The method also includes performing a conversion between a video media file and a bitstream based on the generated filtered samples.

The application discloses a kind of instant information secret communication methods based on video medium, sender will carrier video each frame block and obtain embedding block and non-embedding block, and the instant information to be transmitted is converted into binary data, RS error correction coding is carried out again to obtain binary hidden information, embedding block is embedded to the hidden information, and steganographic compensation is carried out to non-embedding block, and finally obtain video with secret;Receiver is divided to each frame of video with secret and obtains embedding block and non-embedding block, then the embedding block is carried out singular value transformation to extract hidden information, code word is reorganized to obtain information block, then RS error correction decoding is carried out to obtain binary hidden information, and the instant information transmitted is converted to obtain.This kind of instant information secret communication method can hide, extract and other operations to instant information, realize that secret information can still be completely and correctly extracted after subsequent H.264 compression compression encoding after video data with secret uncompressed, and visual quality and embedding capacity can be well maintained.

The invention discloses a video media platform recommendation method based on a distributed robust graph neural network, and the method comprises the steps: constructing a user-video interaction graph based on behavior data, including like, collection and comment, of a user watching a video in a video platform, and carrying out the modeling of the interaction graph through a variational graph auto-encoder, obtaining a low-dimensional embedding vector of the user and the video; thirdly, modeling is carried out on the potential environmental factors; and then, gradually disturbing and reconstructing an embedded vector in combination with a diffusion model, learning a structural relationship embedded in a noise evolution process through a graph neural network, and introducing a distribution robust optimization mechanism to realize personalized video recommendation. According to the method, through fusion graph representation learning, environment modeling and diffusion denoising mechanisms, the influence of data noise and distribution offset on recommendation performance is effectively relieved, and the robustness and recommendation accuracy of a video recommendation system in a complex dynamic environment are improved.

A method implemented by a video coding apparatus includes applying a neural network (NN) filter to an unfiltered sample of a video unit to generate a filtered sample. The NN filter is applied based on a syntax element of the video unit. The method also includes converting between a video media file and a bitstream based on the filtered sample that was generated.

A non-transitory computer-readable storage medium storing a video media bitstream that is encoded by an encoding method, in which a split direction is determined for a coding block to be coded in a triangular prediction mode. A first merge index that indicates first motion information in a merge candidate list for the coding block is determined. A second merge index that indicates second motion information in the merge candidate list for the coding block is determined. A split direction syntax element, a first index syntax element, and a second index syntax element are signaled. A triangular prediction index including a first value indicated by the first index syntax element, a second value indicated by the second index syntax element, and a third value indicated by the split direction syntax element are signaled. The coding block is encoded according to a triangular prediction candidate indicated by the triangular prediction index.

A method of processing video data. The method includes determining that a supplemental enhancement information (SEI) message of a bitstream includes indicators specifying one or more neural network (NN) filter model candidates or selections for a video unit or samples within the video unit, and converting between a video media file comprising the video unit and the bitstream based on the indicators. A corresponding video coding apparatus and non-transitory computer readable medium are also disclosed.

A method of processing video data includes selecting a first dependent quantization (DQ) parameter to be applied to a first video unit within a sequence; selecting a second DQ parameter to be applied to a second video unit within the sequence, wherein the second DQ parameter is different from the first DQ parameter; and converting between a video media file and a bitstream based on the first DQ parameter and the second DQ parameter that were selected. A corresponding video coding apparatus and non-transitory computer readable medium are also disclosed.

This application discloses a method and related apparatus for controlling the projection of video media, relating to the field of software technology. It acquires the vehicle's gear position and the video playback configuration under that gear position, including playback mode and screen configuration. When the gear position changes from P to non-P, the playback mode is video mode, and the screen configuration is multi-screen synchronous playback, a first screen for multi-screen synchronous playback is determined. If the central control screen is the initiating screen and other screens are receiving screens in the first screen, the central control screen is controlled to exit video mode, and a new initiating screen is determined among the other screens in the first screen (excluding the central control screen) to maintain video mode on the other screens. This application effectively solves the experience problem caused by changes in playback mode due to changes in vehicle gear position by shifting the initiating screen. The user's viewing behavior in the car is not disturbed by gear changes, improving the continuity and integrity of the viewing experience.

A method of processing video data includes using a scalability dimension information (SDI) supplemental enhancement information (SEI) message to indicate which primary layers are associated with an auxiliary layer when auxiliary information is present in a bitstream, and converting between a video media file and the bitstream based on the SDI SEI message. A corresponding video coding apparatus and non-transitory computer readable medium are also disclosed.

A mechanism for processing video data is disclosed. The mechanism determines to pad at least one unavailable sample of a video before feeding the video into a process. A conversion is performed between a visual media data and a bitstream based on the padded samples.

Aspects of the present disclosure involve a system comprising a computer-readable storage medium storing a program and a method for generating a summary based on trip information. The program and method include operations for: determining that one or more criteria associated with a user correspond to a trip taken by the user during a given time interval; retrieving a plurality of visual media items generated by a client device of the user during the given time interval; determining location information for the plurality of visual media items; automatically generating a trip graphic to represent the trip based on the plurality of visual media items generated by the user during the given time interval and the determined location information; and causing the trip graphic to be displayed on the client device.

The invention discloses a poster batch generation method and system based on a cartoon video medium, and the method comprises the steps: obtaining the configuration parameters of a one-card video, and the configuration parameters comprise role information and text information; performing image conversion processing on the cartoon video to obtain an image set; identifying the role information of each image in the image set by using a preset model to obtain a role image set; acquiring a poster image set of the cartoon video by utilizing a role position in the role image set and an adding position of the text information; and outputting the poster image set. According to the poster batch generation method and system based on the cartoon video medium, the poster pictures of the cartoon video can be automatically generated, the forms of the poster pictures are various, the production efficiency is greatly improved, the method and system can automatically adapt to the identification process of animation images, and the success rate of the poster is improved.

The present application discloses an audio and video media transmission method and system, which relates to the field of data transmission, including: obtaining video frame data; performing motion analysis on the video frame data to obtain a first region, a second region, and a third region; encoding the video frame data in the first region using an inter-frame prediction coding mode; encoding the video frame data in the second region using a motion compensation coding mode; encoding the video frame data in the third region using an intra-frame prediction coding mode; encoding the motion vector of the second region to obtain fourth encoded data; performing rate control and bitstream encapsulation based on the first, second, and third encoded data to obtain a video stream; merging the video stream and the fourth encoded data to obtain encoded video transmission data, and transmitting the encoded video transmission data. In view of the inflexibility of rate control in video encoding in the prior art, the present application improves the flexibility of rate control.