A computer-implemented method for automating a content release comprising the steps of: determining a content release plan associated with a content release, wherein the content release plan includes a plurality of line items, wherein each line item identifies a content item to be released; parsing the plurality of line items included in the content release plan to build a deployment graph, wherein the deployment graph identifies a plurality of deployment tasks to be executed to complete the content release; executing the content release plan by performing the plurality of deployment tasks according to a schedule identified in the content release plan; and updating the content release plan based on one or more real-time statuses of the plurality of deployment tasks.
H04N 21/262 - Content or additional data distribution scheduling, e.g. sending additional data at off-peak times, updating software modules, calculating the carousel transmission frequency, delaying a video stream transmission or generating play-lists
2.
Physics-Informed Machine Learning Model-Based Corrector for Deformation-Based Fluid Control
ETH Zürich (EIDGENÖSISCHE TECHNISCHE HOCHSCHULE ZÜRICH) (Switzerland)
Inventor
Da Costa De Azevedo, Vinicius
Kim, Byungsoo
Solenthaler, Barbara
Tang, Jingwei
Abstract
A system includes a hardware processor, a machine learning (ML) model-based corrector trained to predict a deformation of a velocity field, and a system memory storing software code. The hardware processor is configured to execute the software code to receive a deformation template and a deformed velocity field produced based on the deformation template, predict, using the ML model-based corrector based on the deformation template and the deformed velocity field, a correction to the deformed velocity field, and correct the deformed velocity field, using the correction, to provide a corrected velocity field. In some implementations, the hardware processor is further configured to execute the software code to advect the corrected velocity field to provide a density field of a corrected simulation of a deformation of a fluid or a viscoelastic material, and produce, using the density field, the corrected simulation.
G06F 30/28 - Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
A system includes a hardware processor, and a system memory storing a software code. The hardware processor executes the software code to receive, from a user, a request related to a digital asset, obtain, from the user, verification data for use in creating a record of ownership of the digital asset, create, using the verification data, the record of ownership of the digital asset in an ownership database, embed a digital watermark into the digital asset to generate a digitally watermarked digital asset, and provide, to the user, the digitally watermarked digital asset. The digital watermark embedded in the digitally watermarked digital asset and the record of ownership of the digital asset in the ownership database enable ownership verification of the digital asset.
A multi-variant content streaming system includes processing hardware and a system memory storing software code. The processing hardware is configured to execute the software code to stream primary content to a plurality of media players, and receive, from a first media player of the plurality of media players, first user interaction data of a first user interacting with the first media player with the primary content. The processing hardware is further configured to execute the software code to generate, using the first user interaction data, first variant content based on the primary content, and stream the first variant content to the first media player while continuing to stream the primary content to media players of the plurality of media players other than the first media player.
H04N 21/472 - End-user interface for requesting content, additional data or services; End-user interface for interacting with content, e.g. for content reservation or setting reminders, for requesting event notification or for manipulating displayed content
H04N 21/45 - Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies
Embodiments provide for parsing a media file using a factor of interest, determining a factor score for the media file, and performing a scored action based on the factor score to provide a media content recommendation to a user/consumer or to content providers. The scored action may include sorting and filtering a media repository, including the media file, which in turn reduces an amount of data needed for a system to provide an objective recommendation to a user, as well as reducing the time and data processing required to provide a recommendation to the user.
Embodiments provide systems and techniques for testing the functionality of a network architecture having multiple redundant internet protocol (IP) networks. An example system includes a first IP network having a first network device, and a second IP network having a second network device. The system also includes a computing device coupled to the first IP network via the first network device and to the second IP network via the second network device. The first network device is configured to selectively forward first packet(s) of the IP media traffic flow to the computing device when the first packet(s) satisfies a first predetermined condition. The second network device is configured to selectively forward second packet(s) of the IP media traffic flow to the computing device when the second packet(s) satisfies a second predetermined condition.
H04L 43/0817 - Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability by checking functioning
H04L 47/32 - Flow control; Congestion control by discarding or delaying data units, e.g. packets or frames
H04L 65/65 - Network streaming protocols, e.g. real-time transport protocol [RTP] or real-time control protocol [RTCP]
7.
UNIQUE PLAYLIST CREATION USING VARIABLE SUBSTITUTION IN VIDEO DELIVERY
In some embodiments, a method sends a playlist that includes links to segments of media content. At least a portion of the links include a variable, and a presentation of the media content that is attributable to a request for the media content is created by insertion of values for variables in the playlist. The values are associated with a first version of the media content or a second version of the media content. The method receives a request for a segment and the request contains information from a link included in the playlist. The information includes data based on a value provided for insertion into the link as a substitution for a variable included in the link. The value is associated with the first version of the media content or the second version of the media content. The segment that corresponds to the link is sent.
H04N 21/262 - Content or additional data distribution scheduling, e.g. sending additional data at off-peak times, updating software modules, calculating the carousel transmission frequency, delaying a video stream transmission or generating play-lists
H04N 21/236 - Assembling of a multiplex stream, e.g. transport stream, by combining a video stream with other content or additional data, e.g. inserting a URL [Uniform Resource Locator ] into a video stream, multiplexing software data into a video stream; Remultiplexing of multiplex streams; Insertion of stuffing bits into the multiplex stream, e.g. to obtain a constant bit-rate; Assembling of a packetised elementary stream
8.
ENVIRONMENTAL FACTOR-BASED SURFACE MAINTENANCE SYSTEM
A system for performing surface maintenance based on one or more detected environmental factors for a surface monitored for maintenance. The system includes a surface monitoring assembly that monitors a surface for one or more environmental factors, such as a traffic pattern in a cleaning application, and the assembly may utilize one or more video cameras focused on a monitored surface along with video analytics software to process captured video to identify an environmental factor, such as heavy traffic, a spill, or the like. When an environmental factor is detected, the assembly passes this data to a maintenance control routine that processes the data to determine whether to vary the operations of an automated robot maintenance system provided in the system. The data can be used to define or modify a location, a time, and/or a parameter of a maintenance task or function (e.g., vacuuming versus wet mopping).
One embodiment of a computer-implemented method for automatically generating warnings in a cloud computing arrangement comprises identifying a plurality of queries residing on a non-federated cloud storage service, at least two queries of the plurality of queries directed to disparate data sources; executing a first query of the at least two queries to produce first extracted information and executing a second query of the at least two queries to produce second extracted information, wherein the second extracted information augments the first extracted information; causing the first extracted information and the second extracted information to be stored in an aggregation table; triggering a processing of the aggregation table to identify a password expiration alert; and causing a generating of a notification associated with the password expiration alert for modification of a password corresponding to a service account.
Embodiments provide for methods, computer program products, and systems to improve media playback comprising receiving a variant stream, identifying respective maximum segment durations for a plurality of different types of client devices that will play media content contained in the variant stream, generating, using the variant stream, a respective playlist for each of the plurality of different types of client devices, wherein the respective playlists each contain different maximum segment durations, and delivering the respective playlists to at least one of the plurality of different types of client devices via a distribution network.
H04N 21/262 - Content or additional data distribution scheduling, e.g. sending additional data at off-peak times, updating software modules, calculating the carousel transmission frequency, delaying a video stream transmission or generating play-lists
H04N 21/2343 - Processing of video elementary streams, e.g. splicing of video streams or manipulating MPEG-4 scene graphs involving reformatting operations of video signals for distribution or compliance with end-user requests or end-user device requirements
H04N 21/239 - Interfacing the upstream path of the transmission network, e.g. prioritizing client requests
H04N 21/845 - Structuring of content, e.g. decomposing content into time segments
11.
Seamless Video Switching by Commercial Off-The-Shelf (COTS) Devices
A switch includes processing hardware and a memory storing software code. The software code is executed to begin receiving a first video stream from a first media source, begin forwarding the first video stream to a video receiver, begin receiving a second video stream from a second media source, receive a switch command to forward the second video stream to the video receiver instead of the first video stream, and detect, within a frame of the first video stream, a switch point in a video packet header of a video packet contained within that frame. The software code begins forwarding, in response to the switch command and at the switch point, the second video stream to the video receiver, and contemporaneously stops forwarding the first video stream to the video receiver, to provide seamless switching from the first video stream to the second video stream.
A transparent display system is provided where broadcast talent (or presenter) can see interactive content, tool palettes, prompts (and the like) as well as their own sketches and annotations, but a viewing audience sees only the broadcast talent and content intended for the viewing audience with the talent's annotation thereof. A transparent scattering screen together with optical filtering or gating of a first optical property of the light (e.g., polarization-based or wavelength-based) is used such that the first property of the light is projected onto the screen so the talent can see the projection, and a camera-side filter blocks the first property of the light so it is not seen by the camera. Simultaneously, a broadcast talent (or presenter) is illuminated by light having properties other than the first property, which allows the talent image to pass through the screen and the camera-side filter allowing the talent to be seen by camera. In some embodiments, a transparent “two-sided” display screen allows people on opposite sides of the screen to see each other, as well as independent 2D or 3D content from each person's side of the screen.
An exoskeleton to support and move a large character costume and to transfer weight of the costume away from a performer's body. The exoskeleton includes a spine including an elongated and rigid vertical member and also includes a waist ring configured to extend about a waist of a performer using the exoskeleton in a spaced apart manner. A lower end of the spine is attached to a rear center portion of the waist ring. The exoskeleton includes a pair of leg assemblies each pivotally coupled on opposite sides of the waist ring. Each of the leg assemblies is spaced apart from a leg of the performer and extends below a foot of the performer, whereby weight is transmitted to a support surface. Each of the leg assemblies may include at least one passive assist mechanism configured to assist the performer in moving the exoskeleton with bipedal locomotion.
A method for training a control system model includes introducing a simulated fault into a software simulation of a physical system and generating emulated sensor data based on the simulated fault, where the emulated sensor data emulates output from one or more sensors of the physical system. The method further includes obtaining output data from a test control system provided with the emulated sensor data, where the test control system emulates a control system of the physical system and tagging the output data with the simulated fault to create training data. The method further includes utilizing the training data to train the control system model, where the control system model is a machine learning model for use with the control system of the physical system during operation of the physical system.
One embodiment of the present invention sets forth a technique for performing face swapping. The technique includes converting a first input image that depicts a first facial identity from a first viewpoint at a first time into a first latent representation and converting a second input image that depicts the first facial identity from a second viewpoint at the first time into a second latent representation. The technique also includes generating, via a first machine learning model, a first output image that depicts a second facial identity from the first viewpoint based on the first latent representation. The technique further includes generating, via the first machine learning model, a second output image that depicts the second facial identity from the second viewpoint based on the second latent representation.
Techniques relating to streaming video are disclosed. These techniques include identifying one or more streaming video sessions for one or more users based on a plurality of events relating to streaming video for the one or more users. The techniques further include storing data for the one or more streaming video sessions in an electronic database, based on the plurality of events, identifying a plurality of metadata relating to the events, and determining, based on a threshold value, a time to store at least a portion of the plurality of metadata in the electronic database, the time occurring after the storing the data for the one or more streaming video sessions. The techniques further include responding to a query for metrics relating to the one or more streaming video sessions by aggregating at least a portion of the stored data.
A system for emotionally enhancing dialogue includes a computing platform having processing hardware and a system memory storing a software code including a predictive model. The processing hardware is configured to execute the software code to receive dialogue data identifying an utterance for use by a digital character in a conversation, analyze, using the dialogue data, an emotionality of the utterance at multiple structural levels of the utterance, and supplement the utterance with one or more emotional attributions, using the predictive model and the emotionality of the utterance at the multiple structural levels, to provide one or more candidate emotionally enhanced utterance(s). The processing hardware further executes the software code to perform an audio validation of the candidate emotionally enhanced utterance(s) to provide a validated emotionally enhanced utterance, and output an emotionally attributed dialogue data providing the validated emotionally enhanced utterance for use by the digital character in the conversation.
G10L 13/08 - Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
G10L 25/63 - Speech or voice analysis techniques not restricted to a single one of groups specially adapted for particular use for comparison or discrimination for estimating an emotional state
A system for producing conversation-driven character animation includes a computing platform having processing hardware and a system memory storing software code, the software code including multiple trained machine learning (ML) models. The processing hardware executes the software code to obtain a conversation understanding feature set describing a present state of a conversation between a digital character and a system user, and to generate an inference, using at least a first trained ML model of the multiple trained ML models and the conversation understanding feature set, the inference including labels describing a predicted next state of a scene within the conversation. The processing hardware further executes the software code to produce, using at least a second trained ML model of the multiple trained ML models and the labels, an animation stream of the digital character participating in the predicted next state of the scene within the conversation.
Techniques for generating translated audio output based on media content are disclosed. Text is accessed corresponding to media content. One or more untranslated mouth shape indicia are determined based on the text. The text is parsed into one or more text chunks when one or more dubbing parameters are met. The parsed text is translated from a first spoken language to a second spoken language. One or more translated mouth shape indicia are determined. The one or more translated mouth shape indicia and the one or more untranslated mouth shape indicia are compared based on a predetermined tolerance threshold. A translated audio output is generated based on the translated text.
A media enhancement system includes an augmented reality (AR) device having a display, processing hardware, and a memory storing software code. The processing hardware executes the software code to monitor media content including a sequence of moving images displayed on a display screen separate from the AR device, receive playhead data indicating a playhead state of a media playout device playing out the media content, and detect, based on monitoring the media content, one or more image(s) in the sequence of moving images as one or more anchor image(s). The software code is further executed to obtain, using the anchor image(s), one or more AR effect(s) associated with the anchor image(s), and render, based on the playhead data, the AR effect(s) on the display of the AR device, wherein the AR effect(s) is/are spatially and temporally aligned with the sequence of moving images being displayed on the display screen.
A system includes an augmented reality (AR) device having a first display, processing hardware, and a memory storing software code. The processing hardware is configured to execute the software code to monitor media content including a sequence of moving images displayed on a second display separate from the AR device and controlled by a media player device, detect, based on monitoring the media content, an image in the sequence of moving images for enhancement by one or more AR effects. The processing hardware further executes the software code to render the one or more AR effects on the first display, and transmit, contemporaneously with rendering the one or more AR effects on the first display, a signal configured to pause or loop the playing of the media content on the second display.
Systems and methods for training a neural network to predict states of a robotics device are disclosed. Robotics data is received for a robotics device, including indications of a set of components, a digital simulation of the robotics device, and measurement data received from a sensor associated with the robotics device. The set of components includes an actuator and a structural element. A training dataset is generated using the received robotics data. Generating the training dataset includes comparing the measurement data with simulated measurement data based on the digital simulation. A neural network is trained using the generated training dataset to modify the digital simulation of the robotics device to predict a state of the robotics device, such as a position, motion, electrical quantity, or other. When trained, the neural network is applied to predict states of the robotics device or a different robotics device.
In some embodiments, a method receives input data to calculate an effect of a variable on a group for a plurality of methods. Methods in the plurality of methods calculate the effect of the variable for the input data using different logic. A plurality of sub-weights for methods in the plurality of methods are generated. The sub-weights are generated based on a balance metric, a dissimilarity metric, and a reliability metric. The method combines the plurality of sub-weights for methods in the plurality of methods to generate a final weight for the methods. The respective final weight is applied to an intermediate result from a respective method in the plurality of methods to generate a weighted intermediate result for the method. The method combines weighted intermediate results for the plurality of methods to generate a final result for the effect of the variable.
A system includes an augmented reality (AR) device having a display, processing hardware, and a memory storing software code. The processing hardware executes the software code to monitor media content including a sequence of images displayed on a display screen separate from the AR device, detect, based on monitoring the media content, an image in the sequence of images as an anchor image, and obtain, using the anchor image, one or more AR effect(s) associated with the anchor image. The processing hardware further executes the software code to determine a position and orientation of the AR device in relation to the display screen, and render, based on that position and orientation, the AR effect(s) on the display of the AR device, where the AR effect(s) include at least one intermediate scale AR effect having a scale intermediate between a display screen scale AR effect and a real-world scale AR effect.
SYSTEMS AND METHODS TO PRODUCE A PHYSICAL ARTICLE THAT PROVIDES ADMISSION AUTHORIZATION TO AN EVENT WHICH CORRELATES TO A DIGITAL ASSET FROM A TEMPORARY WALLET
Systems and methods to produce a physical article that provides admission authorization to an event and correlated digital asset from a temporary wallet are disclosed. Exemplary implementations may: establish one or more temporary wallets; effectuate a smart contract that causes the temporary wallets to hold digital assets, where individual digital assets are correlated with event information; generate one or more wallet machine-readable mediums that represent the temporary wallets; generate one or more asset machine-readable mediums that represent the digital assets; cause an output device to output the wallet machine-readable mediums and the wallet machine-readable mediums on physical articles that provide admission authorizations events or locations; upon user devices scanning the wallet machine-readable mediums, the users may be enabled to establish digital wallets or access digital wallets such that the digital asset is transferred from the temporary wallets to the digital wallets.
G06Q 20/40 - Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check of credit lines or negative lists
G06Q 20/36 - Payment architectures, schemes or protocols characterised by the use of specific devices using electronic wallets or electronic money safes
A system includes a computing platform having processing hardware, and a memory storing software code. The software code is executed to receive digital content indexed to a timeline, receive insertion data identifying a timecode of the timeline, and encode the digital content using the insertion data to provide segmented content having a segment boundary at the timecode, and first and second segments adjoining the segment boundary, wherein the first segment precedes, and the second segment succeeds, the segment boundary. The software code also re-processes the first and second segments to apply a fade-out within or to the first segment and a fade-in within or to the second segment, wherein re-processing the first and second segments provides encoded segments having the segment boundary configured as an insertion point for supplemental content.
A system includes processing hardware and a memory storing software code. The processing hardware executes the software code to receive automation data for media content having a default playback experience, analyze, using the automation data, at least one parameter of the media content, and generate, based on the analyzing, one or more automation instruction(s) for at least one portion(s) of the media content. The automation instruction(s) include at least one of: one or more bounding timestamps of the media content portion(s), an increased or reduced playback speed for the media content portion(s) relative to the default playback experience, or a variable playback speed for the media content portion(s). The software code is further executed to outputs the automation instruction(s) to a media delivery platform configured to distribute and control the quality of the media content or to a media player configured to automate playback of the media content.
Systems and methods to transfer a digital asset to a digital wallet from a temporary wallet are disclosed. Exemplary implementations may: establish a temporary wallet; effectuate a smart contract that causes the temporary wallet to hold a digital asset; generate a machine-readable medium that represent the temporary wallet; cause an output device to output the machine-readable medium on a physical article; upon a user device scanning the machine-readable medium, the user may be enabled to establish digital wallets or access digital wallets such that the digital asset is transferred from the temporary wallet to the digital wallet.
G06Q 20/10 - Payment architectures specially adapted for home banking systems
G06Q 20/36 - Payment architectures, schemes or protocols characterised by the use of specific devices using electronic wallets or electronic money safes
A system for creating accessibility enhanced content includes a computing platform having processing hardware and a system memory storing a software code and a machine learning (ML) model, the software code providing a graphical user interface (GUI). The processing hardware executes the software code to identify a user of the system, obtain a user profile of the user, obtain, from one or more application(s) utilized by the user, activity data relating to use of the application(s) by the user, and modify, using the user profile and the activity data, one or more node weights of the ML model to provide a tuned ML model. The processing hardware further executes the software code to infer, using the tuned ML model, at least one action for advancing a development of the user and output to the user, using the UI, a recommendation for performing the at least one action by the user.
Systems and methods to determine content for a narrative based on user input related to the narrative are disclosed. Exemplary implementations may: effectuate presentation of a narrative via client computing platforms associated with users; obtain, user inputs in relation to the active content; initiate a smart contract configured to receive the user inputs; initiate a smart contract configured to determine whether the user inputs include one or more of the trigger events; initiate a smart contract configured to determine the inactive content for presentation upon determination that one or more of the trigger events occurred; initiate a smart contract configured to transmit an indication to the one or more processors to effectuate presentation of the inactive content; and initiate a smart contract configured to generate and execute a set of instructions to record the reclassification and the active content on the decentralized ledger.
G06Q 20/12 - Payment architectures specially adapted for electronic shopping systems
G06Q 20/36 - Payment architectures, schemes or protocols characterised by the use of specific devices using electronic wallets or electronic money safes
31.
SYSTEMS AND METHODS TO GENERATE A COMPOSITE REPRESENTATION OF A DIGITAL WALLET BASED ON DIGITAL ASSETS INDICATED BY THE DIGITAL WALLET
Systems and methods to generate a composite representation of a digital wallet based on digital assets indicated by the digital wallet are disclosed. Exemplary implementations may: determine a set of digital assets indicated as owned by digital wallets; determine, based on the set of digital assets, the correlated entities; analyze the correlated entities to determine a composite representation of the entities or adjustments to the composite representation; mint a composite digital asset correlated with the composite representation; and update, upon determination of the adjustments to the composite representation, the composite representation in accordance with the adjustments.
G06Q 20/36 - Payment architectures, schemes or protocols characterised by the use of specific devices using electronic wallets or electronic money safes
32.
SYSTEMS AND METHODS TO IMPLEMENT PHYSICAL AND/OR BEHAVIORAL ATTRIBUTES THAT DEFINE A PERSONALITY ENTITY ASSOCIATED WITH A DIGITAL ASSET VIA A PHYSICAL OBJECT
Systems and methods to implement physical and/or behavioral attributes that define a personality entity associated with a digital asset via a physical object are disclosed. Exemplary implementations may: determine physical and/or behavioral attributes associated with digital assets held by a digital wallet; receive output signals from sensors; determine, based on the output signals, interaction information; determine, based on the decentralized ledger and the address, changes in ownership of the digital assets; determine and effectuate adjustments to the physical and/or behavioral attributes based on the interaction information and the changes in ownership; transmit the adjustments to the physical object such that the physical object exemplifies the adjustments; receive an indication that the private portable device decoupled from the hardware coupler; and initiate, responsive to the indication, a wrap smart contract that is configured to link ownership rights to the digital assets in the digital wallet into a packaged digital asset.
G06Q 20/36 - Payment architectures, schemes or protocols characterised by the use of specific devices using electronic wallets or electronic money safes
The exemplary embodiments relate to converting Standard Dynamic Range (SDR) content to High Dynamic Range (HDR) content using a machine learning system. In some embodiments, the neural network is trained to convert an input SDR image into an HDR image using the encoded representation of a training SDR image and a training HDR image. In other embodiments, the neural network is trained to convert an input SDR image into an HDR image using a predefined set of color grading actions and the training images.
A system includes a computing platform having processing hardware, and a memory storing software code. The software code is executed to receive content having a sequence of content segments, and marker data identifying a location within the sequence, identify, using the content and the marker data, segment boundaries of a content segment containing the location, determine, using the location and the segment boundaries, whether the location is situated within a predetermined interval of one of the segment boundaries, and re-encode a subsection of the sequence to produce a new segment boundary at the location. When the location is not situated within the predetermined interval, the subsection of the sequence includes the content segment containing the location. When the location is situated within the predetermined interval, the subsection of the sequence includes the content segment containing the location and a content segment adjoining the content segment containing the location.
A configuration-based system and method for handling transient data in complex systems includes a method for testing a Host App running on a user device, the Host App communicating with at least one API server using a Host App API Request, the Host App API Request including a Host App API configuration request URL, the Host App API configuration request URL being part of a Host App API Config File, the method includes selectively hydrating the Host App API Config File by selectively replacing at least one of the Host App API configuration request URLs corresponding to a query in a Test Config File with a hydrated Host App API configuration request URL, thereby causing the Host App API Request to route through Test Logic which provides a response to the Host App Request based on desired test behavior defined in the Test Config File.
In some embodiments, a method receives usage data that is based on a delivery of content by a plurality of content delivery network entities. A first value of a selection parameter is used to determine whether to select a content delivery network entity from the plurality of content delivery network entities to process a first request for content. The method allocates the usage data in a first distribution to the plurality of content delivery network entities. The allocating does not use a condition to determine the first distribution. The usage data is allocated in a second distribution to the plurality of content delivery network entities. The allocating uses the condition to determine the second distribution. The method adjusts the first value of the selection parameter to a second value based on the first distribution and the second distribution.
A behavior detection platform may obtain video data of an environment including a guest and may determine a set of normal guest behaviors associated with the environment. The platform may analyze, using a first machine learning model, the video data to identify features of the guest and analyze, using a second machine learning model, the video data and the features to determine an actual guest behavior of the guest. The platform may predict, using the second machine learning model, a predicted guest behavior of the guest based on the actual guest behavior. The behavior detection platform may determine that at least one of the actual guest behavior or the predicted guest behavior is not included in the set of normal guest behaviors. The platform may cause assistance to be provided to the guest based on determining that the actual guest behavior or the predicted guest behavior are not included.
A control system, and associated control method, that processes aircraft tracking data to determine precise location information for aircraft in the vicinity of an outdoor show that includes laser projectors. The aircraft location information for each aircraft is processed along with heading and speed data to generate a set of laser control data, which is communicated via a monitoring interface to a laser projector operator for use in operating the laser projector. The control system determines the scan field into which the laser projector may project its light during the show, and this scan field is provided, e.g., visually in the monitoring interface, to the laser projector operator along with the sets of laser control data for each aircraft. The laser projector operator may then terminate or continue to operate the laser projector based on this very precise information related to the scan field and location of aircraft.
A system includes a hardware processor and a memory storing software code. The hardware processor executes the software code to determine how many users are present at a location at a starting time of an NFT acquisition event, identify, based on how many users are present, a collaborative activity for the users, the collaborative activity requiring relinquishment of a plurality of NFTs, and communicate the collaborative activity to the users. The hardware processor further executes the software code to verify that the collaborative activity has been completed, and provide, in response to verifying, one or more new NFT(s) for distribution to at least one of the users, wherein (i) a value associated with the new NFT(s) exceeds a value of the NFTs relinquished, or (ii) a number of the NFTs relinquished exceeds a number of the new NFT(s).
G06Q 20/06 - Private payment circuits, e.g. involving electronic currency used only among participants of a common payment scheme
G06Q 20/40 - Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check of credit lines or negative lists
40.
Interactive gameplay system with play augmented by player-selected customization preferences
A system for providing user-driven customization and enhanced personalization of interactive experiences. The system includes data storage for storing player profiles, with each including customization preferences useful in enhancing or generating one of the interactive experiences. The system includes a gameplay space adapted to provide an interactive experience, which includes one or more interactive elements. The system includes a gameplay device configured to be worn or carried by a player. A detection device detects a presence of the player in the gameplay space and obtains a unique identifier for the gameplay device. The system includes a controller retrieving a set of the customization preferences in one of the player profiles associated with the identifier. During system operations, the interactive experience is provided to the player with interactive elements generated based on the retrieved set of customization preferences, whereby the player can affect and enhance their interactive experience in real time.
A63F 13/212 - Input arrangements for video game devices characterised by their sensors, purposes or types using sensors worn by the player, e.g. for measuring heart beat or leg activity
A63F 13/2145 - Input arrangements for video game devices characterised by their sensors, purposes or types for locating contacts on a surface, e.g. floor mats or touch pads the surface being also a display device, e.g. touch screens
A63F 13/235 - Input arrangements for video game devices for interfacing with the game device, e.g. specific interfaces between game controller and console using a wireless connection, e.g. infrared or piconet
A63F 13/79 - Game security or game management aspects involving player-related data, e.g. identities, accounts, preferences or play histories
A system for assessing performer density includes a computing platform having processing hardware and a memory storing software code. The processing hardware is configured to execute the software code to receive content and content data, the content data identifying one or more performer(s) depicted or referenced in the content, and determine, using the content and the content data, one or more segment(s) of the content in which the performer(s) is/are depicted or referenced. The processing hardware is further configured to execute the software code to infer, for each determined segment(s) of the content, a respective importance of the performer(s) in a respective context of each determined segment(s), and calculate, based on the determined segment(s) of the content and the respective importance of the performer(s), a respective density score of each performer with respect to the content.
A system includes a hardware processor and a memory storing software code. The hardware processor executes the software code to receive a request from a user to link a non-fungible token (NFT) with a user account of the user, verify, in response to receiving the request, ownership of the NFT by the user, and link, in response to verifying, the NFT to the user account. The hardware processor is further configured to execute the software code to modify, based on at least one of the content viewing history of the user or a release date of a content available to the user, one or more attributes of the NFT.
ETH Zurich (EIDGENÖSSISCHE TECHNISCHE HOCHSCHULE ZURICH) (Switzerland)
Inventor
Schroers, Christopher Richard
Azevedo, Roberto Gerson De Albuquerque
Gregory, Nicholas David
Xue, Yuanyi
Labrozzi, Scott
Djelouah, Abdelaziz
Abstract
A system includes a machine learning (ML) model-based video downsampler configured to receive an input video sequence having a first display resolution, and to map the input video sequence to a lower resolution video sequence having a second display resolution lower than the first display resolution. The system also includes a neural network-based (NN-based) proxy video codec configured to transform the lower resolution video sequence into a decoded proxy bitstream. In addition, the system includes an upsampler configured to produce an output video sequence using the decoded proxy bitstream.
H04N 19/147 - Data rate or code amount at the encoder output according to rate distortion criteria
H04N 19/132 - Sampling, masking or truncation of coding units, e.g. adaptive resampling, frame skipping, frame interpolation or high-frequency transform coefficient masking
H04N 19/184 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being bits, e.g. of the compressed video stream
ETH Zürich (Eidgenössische Technische Hochschule Zürich) (Switzerland)
Inventor
Kansy, Manuel Jakob
Raël, Anton Julien
Naruniec, Jacek Krzysztof
Schroers, Christopher Richard
Weber, Romann Matthew
Abstract
One embodiment of the present invention sets forth a technique for performing identity-preserving image generation. The technique includes converting an identity image depicting a facial identity into an identity embedding. The technique further includes generating a combined embedding based on the identity embedding and a diffusion iteration identifier. The technique further includes converting, using a neural network and based on the combined embedding, a first input image that includes first noise into a first predicted image depicting one or more facial features that include one or more first facial identity features, wherein the one or more first facial identity features correspond to one or more respective second facial identity features of the identity image and are based at least on the identity embedding.
A system includes a hardware processor, and a system memory storing a software code and a machine learning (ML) model trained to apply a stylization to an image. The hardware processor executes the software code to receive a first sequence of images and style data describing a desired stylization of content depicted by the first sequence of images. The hardware processor further executes the software code to stylize the content, using the ML model, to provide a stylized content having the desired stylization, wherein stylizing includes applying an exponential moving average (EMA) temporal smoothing algorithm to sequential image pairs of the first sequence of images to generate a second sequence of images providing a depiction of the content having the desired stylization, and output the stylized content having the desired stylization.
G06F 30/28 - Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
46.
Dynamic Authorization Rule Stacking and Routing Across Multiple Systems
A system includes a hardware processor that executes a software code to receive an authorization request on behalf of a user for a stacked resource including resources offered separately by multiple resource providers, determine resource provider computers associated with the stacked resource, and send a look-up request including an electronic identity of the user to those computers, where the electronic identity is used as a look-up key for determining user attribute(s) of the user. The software code further receives the user attribute(s) from the resource provider computers, generates an accumulated access profile of the user based on the user attribute(s), applies the profile to a rules engine to determine a stacked access result, and routes the authorization request and the stacked access result to one of the resource provider computers, where that computer completes an authorization process for access to the stacked resource based on the stacked access result.
ETH Zürich (Eidgenössische Technische Hochschule Zürich) (Switzerland)
Inventor
Naruniec, Jacek Krzysztof
Kansy, Manuel Jakob
Mignone, Graziana
Schroers, Christopher Richard
Weber, Romann Matthew
Abstract
One embodiment of the present invention sets forth a technique for performing face swapping. The technique includes generating a latent representation of a first facial identity included in an input image. The technique further includes identifying a first identity-specific neural network layer associated with a second facial identity from a plurality of identity-specific neural network layers, wherein each neural network layer included in the plurality of identity-specific neural network layers is associated with a different facial identity. The technique further includes executing the first identity-specific neural network layer and one or more other neural network layers to generate one or more decoder input values corresponding to the latent representation. The technique further includes executing a decoder neural network that converts the one or more decoder input values into an output image depicting the second facial identity.
ETH Zürich (Eidgenössische Technische Hochschule Zürich) (Switzerland)
Inventor
Djelouah, Abdelaziz
Bernasconi, Michael Yves
Salehi, Farnood
Schroers, Christopher Richard
Abstract
Techniques are disclosed for resampling images. In some embodiments, a resampling model includes (1) one or more feature extraction layers that extract features from an input image and a degradation map; (2) one or more resampling layers that generate warped features from the extracted features and a warp grid; and (3) one or more prediction layers that generate, from the warped features, an output image or resampling kernels that can be applied to the input image to generate an output image. In some embodiments, the resampling model can be trained by applying degradation maps to output images in a training data set to generate corresponding input images, and training the resampling model using the input images and the corresponding output images.
Techniques for automatically placing and manipulating virtual objects of augmented reality (AR) and mixed reality (MR) simulations are described. One technique includes obtaining an indication of at least one virtual object available for placing within a real-world environment during an AR simulation or a MR simulation. A first representation of the real-world environment is generated, based on a scan of the real-world environment. At least one second representation of the real-world environment is generated from the first representation. A match is determined between the at least one virtual object and at least one available space within the real-world environment, based at least in part on evaluating the at least one virtual object and the at least one second representation with a machine learning model(s). The at least one virtual object is rendered on a computing device, based on the match.
An example system includes a database having a plurality of data structures, each of the plurality of data structures associated with a different business rule and one or more provider identifications, and a processor configured to receive one or more consumer identifications and one of the one or more provider identifications, search the plurality of data structures for one or more data structures associated with the one of the one or more provider identifications to identify authorized data structures, in the authorized data structures, determine entitlements associated with the one or more consumer identifications to identify consumer entitlements, generate a list of the consumer entitlements, and transmit the list of the consumer entitlements in response to the receiving.
Embodiments herein describe a CDN where anycast routing is used to identify a load balancer for selecting a cache in the CDN to use to deliver a requested object to a user. In one embodiment, the user performs a DNS lookup to identify an anycast IP address for a plurality of load balancers in the CDN. The user can then initiate anycast routing using the anycast IP address to automatically identify a load balancer. Once the identified balancer selects the cache, the load balancer can close the anycast connection with the user device and use a redirect to provide the user device with a unicast path to the selected cache. The user device can then establish a unicast connection with the cache to retrieve (e.g., stream) the object.
H04L 67/1036 - Load balancing of requests to servers for services different from user content provisioning, e.g. load balancing across domain name servers
H04L 61/4511 - Network directories; Name-to-address mapping using standardised directory access protocols using domain name system [DNS]
H04L 67/568 - Storing data temporarily at an intermediate stage, e.g. caching
H04L 67/02 - Protocols based on web technology, e.g. hypertext transfer protocol [HTTP]
H04L 67/10 - Protocols in which an application is distributed across nodes in the network
H04L 67/1004 - Server selection for load balancing
H04L 67/101 - Server selection for load balancing based on network conditions
H04L 67/1021 - Server selection for load balancing based on client or server locations
Techniques are disclosed for denoising videos. In some embodiments, video frames are denoised using a denoising model that includes an encoder-decoder architecture and attention modules. During training of the denoising model, the attention modules learn weightings to upweight certain dimensions of input features to help pixel registration, remove ghosting artifacts, and improve temporal consistency when the frames of a video are being denoised. The denoising model can also be used to train a student denoising model that has a same architecture as, but is smaller and faster than, the denoising model. After training, noisy video frames can be input into the denoising model and/or the student denoising model to generate corresponding denoised video frames.
Certain aspects of the present disclosure provide techniques for adaptive sampling for rendering using deep learning. This includes receiving, at a sampler in a rendering pipeline, a plurality of rendered pixel data, wherein the sampler includes a first machine learning (ML) model. It further includes generating a sampling map for the rendering pipeline using the first ML model and the plurality of rendered pixel data, including predicting a plurality of pixel values in the sampling map based on a generated distribution of pixel values. It further includes rendering an image using the sampler, the sampling map, and a denoiser in the rendering pipeline.
A computer-implemented method of changing a face within an output image or video frame that includes: receiving an input image that includes a face presenting a facial expression in a pose; processing the image with a neural network encoder to generate a latent space point that is an encoded representation of the image; decoding the latent space point to generate an initial output image in accordance with a desired facial identity but with the facial expression and pose of the face in the input image; identifying a feature of the facial expression in the initial output image to edit; applying an adjustment vector to a latent space point corresponding to the initial output image to generate an adjusted latent space point; and decoding the adjusted latent space point to generate an adjusted output image in accordance with the desired facial identity but with the facial expression and pose of the face in the input image altered in accordance with the adjustment vector
A computer-implemented method of changing a face within an output image or video frame includes: receiving an input image that includes a face presenting a facial expression in a pose; separately encoding different portions of the image by, for each separately encoded portion, generating a latent space point of the portion, thereby generating a plurality of multi-dimensional vectors where each multi-dimensional vector is an encoded representation of a different portion of the input image; concatenating the plurality of multi-dimensional vectors into a combined latent space vector; and decoding the combined latent space vector to generate the output image in accordance with a desired facial identity but with the facial expression and pose of the face in the input image
Embodiments provide systems and techniques for a network architecture design for a converged media facility. An example system includes a plurality of internet protocol (IP) networks, where each IP network is configured to provide a different service for the media facility. At least one of the IP networks is a production media network configured to route multicast media traffic. The system also includes a plurality of first network devices coupled to the plurality of IP networks and a controller configured to manage at least one IP traffic flow within at least one of the plurality of IP networks.
A robot design and control system including a procedural animation engine and a graphical animation editor that enable animators to author stylized walking gaits achievable by physical robotic characters, including bipedal robots of varying design. The animation engine generates dynamically feasible reference trajectories for omnidirectional walking given a desired walking velocity that may be input from a joystick or an artificial intelligence (AI) planner. This allows a legged robot to walk along an arbitrary path while expressing a custom animation “style,” e.g., a happy walk, a sneaky walk, or other manner of walking. The stylized walking motion or gait is generalized by the animation engine from a small number of animation samples that are defined at key walking velocities and tracked using a whole-body controller. The set of samples that are used as input to define a walking style is authored by an animator using the animation editor.
G06T 13/40 - 3D [Three Dimensional] animation of characters, e.g. humans, animals or virtual beings
B25J 5/00 - Manipulators mounted on wheels or on carriages
G06F 3/04847 - Interaction techniques to control parameter settings, e.g. interaction with sliders or dials
G06F 3/04815 - Interaction with a metaphor-based environment or interaction object displayed as three-dimensional, e.g. changing the user viewpoint with respect to the environment or object
G06F 30/20 - Design optimisation, verification or simulation
An entertainment system for a vehicle is disclosed. In one embodiment, the entertainment system includes an at least partially enclosed structure configured to receive an on-road vehicle, where the structure defines a path along which the on-road vehicle is configured to move autonomously; a content output system configured to generate content based on one or more characteristics of the on-road vehicle; and a guide system configured to guide the on-road vehicle along the path.
B60W 10/30 - Conjoint control of vehicle sub-units of different type or different function including control of auxiliary equipment, e.g. air-conditioning compressors or oil pumps
B60W 10/18 - Conjoint control of vehicle sub-units of different type or different function including control of braking systems
B60W 10/20 - Conjoint control of vehicle sub-units of different type or different function including control of steering systems
Certain aspects of the present disclosure provide techniques for automatic compute environment scheduling using machine learning (ML). This includes identifying a first compute resource among a plurality of compute resources operating in a compute infrastructure, where the first compute resource is in a first operational state. It further includes determining, based on comparing a first time with a compute resources schedule generated using an ML model, that the first compute resource should be placed in a second operational state different from the first operational state. It further includes determining whether the compute resources schedule should be disregarded, and either (1) in response to determining that the compute resources schedule should not be disregarded, placing the first compute resource in the second operational state, or (2) in response to determining that the compute resources schedule should be disregarded, allowing the first compute resource to remain in the first operational state.
In some implementations, a controller may cause a robotic arm connector, of the robotic arm, to be connected to a first compartment connector of a passenger compartment of an amusement ride. The controller may cause a second compartment connector, of the passenger compartment, to be disconnected from a first structure connector of a first support structure. The controller may transport the passenger compartment from the first support structure to a second support structure after causing the second compartment connector to be disconnected from the first structure connector.
The present disclosure relates to an entertainment system for a vehicle. In one embodiment, the entertainment system includes a display external to the vehicle and viewable by an occupant of the vehicle. The entertainment system includes a communication module to receive occupant preferences from an occupant, and a processor configured to generate or modify content to be presented on the display based on the occupant preferences.
According to one implementation, a system includes a hardware processor and a memory storing software code. The hardware processor executes the software code to receive ownership data identifying a unique digital identifier conferring ownership, by a user, of an asset awarded by a first entity, confirm, using the unique digital identifier, ownership of the asset by the user, and verify, using the unique digital identifier, that a second entity is an authorized asset redemption partner of the first entity. The hardware processor further executes the software code to enable, using the unique digital identifier and in response to confirming and verifying, the user to redeem the asset awarded by the first entity from the second entity via transfer from a first digital wallet linked to the first entity to a second digital wallet linked to the second entity, either automatically or in response to an authentication performed by the user.
A system includes a computing platform having processing hardware and a memory storing an asset library and software code including a recommendation engine. The processing hardware executes the software code to receive, from a user, group generation data and a preferences profile of the user, the group generation data identifying a group including the user and another user, and to send, to the other user, an invitation to join the group. The processing hardware also executes the software code to receive, from the other user, one of an acceptance or a refusal of the invitation, obtain, in response to receiving the acceptance, a preferences profile of the other user, generate, using the preferences profiles of the user and the other user, a group preferences profile for the group, and identify, using the recommendation engine and the group preferences profile, one or more assets in the asset library for the group.
H04N 21/2668 - Creating a channel for a dedicated end-user group, e.g. by inserting targeted commercials into a video stream based on end-user profiles
H04N 21/25 - Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication or learning user preferences for recommending movies
64.
REAL-TIME VIBRATION-SUPPRESSION CONTROL FOR ROBOTIC SYSTEMS
In one example, a robotic system is disclosed that includes a plurality of components coupled together, a plurality of motors operable to move the plurality of components, a controller in electrical communication with the plurality of motors to generate control signals to actuate movement of the plurality of components, wherein the controller is configured to: receive a first set of control signals operative to generate a defined motion for the plurality of components, analyze the first set of control signals to determine a second set of control signals operative to define a retargeted motion for the plurality of components, wherein the retargeted motion suppresses vibrations of the plurality of components as compared to the defined motion, and provide the second set of control signals to the plurality of motors to actuate the retargeted motion by the plurality of components.
A content delivery system includes a computing platform having processing hardware and a system memory storing software code, a user account database, and a content database. The processing hardware executes the software code to receive, from a user device utilized by a user, a request for access to a first content of a plurality of contents, the request including an identifier of a sponsor of the request, to verify, using the identifier, an authorization status of the sponsor, and transmit to the user device, based on verification of the authorization status of the sponsor, content access data enabling access to the first content by the user. The processing further executes the software code to detect that the user is accessing the first content, and to offer, in response to detecting that access, an opportunity to the user to obtain access to the plurality of contents.
H04N 21/442 - Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed or the storage space available from the internal hard disk
H04N 21/61 - Network physical structure; Signal processing
H04N 21/258 - Client or end-user data management, e.g. managing client capabilities, user preferences or demographics or processing of multiple end-users preferences to derive collaborative data
66.
BEHAVIOR-BASED COMPUTER VISION MODEL FOR CONTENT SELECTION
Various embodiments set forth systems and techniques for evaluating media content items. The techniques include receiving visual feedback associated with one or more audience members viewing a first media content item; analyzing the visual feedback to generate one or more emotion signals based on the visual feedback; and generating a set of features associated with the one or more audience members viewing the first media content item based on the one or more emotion signals.
ETH Zurich (Eidgenossische Technische Hochschule Zurich) (Switzerland)
Inventor
Zhang, Yang
Song, Mingyang
Aydin, Tunc Ozan
Schroers, Christopher Richard
Abstract
Techniques are disclosed for enhancing videos using a machine learning model that is a temporally-consistent transformer model. The machine learning model processes blocks of frames of a video in which the temporally first input video frame of each block of frames is a temporally second to last output video frame of a previous block of frames. After the machine learning model is trained, blocks of video frames, or features extracted from the video frames, can be warped using an optical flow technique and transformed using a wavelet transform technique. The transformed video frames are concatenated along a channel dimension and input into the machine learning model that generates corresponding processed video frames.
Various embodiments for recommending and displaying content are disclosed. In one example, a method is disclosed that includes generating a first set of content portions from a plurality of content portions based on a plurality of metadata tags in a user profile, wherein the content portions are a subset of frames of a content item; displaying the first set of content portions in an interface provided to a user device corresponding to the user profile; filtering the first set of content portions based on one or more filter inputs to generate a filtered set of content portions; and transmitting the at least one content portion in the filtered set of content portions to the user device for presentation to the user.
H04N 21/472 - End-user interface for requesting content, additional data or services; End-user interface for interacting with content, e.g. for content reservation or setting reminders, for requesting event notification or for manipulating displayed content
H04N 21/235 - Processing of additional data, e.g. scrambling of additional data or processing content descriptors
A display system for a providing a three-dimensional effect with controlled reflections is disclosed. In one embodiment, the display system includes a display; and a beam splitter positioned between the display and the viewing location. The beam splitter transmits or reflects light based on a characteristic of the light. A physical prop is positioned between the beam splitter and the viewing location. The display is oriented to emit light toward the beam splitter and displays an image at least partially viewable in the viewing location through the beam splitter, such that the image is substantially un-attenuated by the bean splitter. Light reflected from the physical prop and further reflected by the beam splitter toward the viewing location is at least partially viewable in the viewing location. The display system is suitable to enable a substantially full brightness of the image to be viewable in the viewing location.
G02B 30/52 - Optical systems or apparatus for producing three-dimensional [3D] effects, e.g. stereoscopic images the image being built up from image elements distributed over a 3D volume, e.g. voxels the 3D volume being constructed from a stack or sequence of 2D planes, e.g. depth sampling systems
G02B 30/56 - Optical systems or apparatus for producing three-dimensional [3D] effects, e.g. stereoscopic images the image being built up from image elements distributed over a 3D volume, e.g. voxels by projecting aerial or floating images
G02B 27/14 - Beam splitting or combining systems operating by reflection only
H04N 13/337 - Displays for viewing with the aid of special glasses or head-mounted displays [HMD] using polarisation multiplexing
70.
Data Object Classification Using an Optimized Neural Network
ETH Zürich (EIDGENÖSSISCHE TECHNISCHE HOCHSCHULE ZÜRICH) (Switzerland)
Inventor
Riemenschneider, Hayko Jochen Wilhelm
Helminger, Leonhard Markus
Schroers, Christopher Richard
Djelouah, Abdelaziz
Abstract
A system includes a computing platform having a hardware processor and a memory storing a software code and a neural network (NN) having multiple layers including a last activation layer and a loss layer. The hardware processor executes the software code to identify different combinations of layers for testing the NN, each combination including candidate function(s) for the last activation layer and candidate function(s) for the loss layer. For each different combination, the software code configures the NN based on the combination, inputs, into the configured NN, a training dataset including multiple data objects, receives, from the configured NN, a classification of the data objects, and generates a performance assessment for the combination based on the classification. The software code determines a preferred combination of layers for the NN including selected candidate functions for the last activation layer and the loss layer, based on a comparison of the performance assessments.
G06V 10/764 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
G06V 10/774 - Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
A system includes a computing platform having processing hardware, and a systems memory storing a software code. The processing hardware is configured to execute the software code to receive content including an image having multiple image regions, determine boundaries of each of the image regions to identify multiple bounded image regions, identify, within each of the bounded image regions, one or more local features and one or more global features, and identify, within each of the hounded image regions, another one or more local features based on a comparison with corresponding local features identified in each of one or more other bounded image regions. The processing hardware is further configured to execute the software code to annotate each of the bounded image regions using its respective one or more local features, its other one or more local features, and its one or more global features, to provide annotated content.
G06V 20/70 - Labelling scene content, e.g. deriving syntactic or semantic representations
G06V 10/42 - Global feature extraction by analysis of the whole pattern, e.g. using frequency domain transformations or autocorrelation
G06V 10/44 - Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
G06V 10/70 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning
G06V 20/30 - Scenes; Scene-specific elements in albums, collections or shared content, e.g. social network photos or video
G06V 30/414 - Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text
G06V 30/42 - Document-oriented image-based pattern recognition based on the type of document
One embodiment of the present invention sets forth a technique for performing remastering of video content. The technique includes determining a first input frame corresponding to a first frame included in a first video and a first target frame corresponding to a second frame included in a second video based on one or more alignments between the first frame and the second frame. The technique also includes executing a machine learning model to convert the first input frame into a first output frame. The technique further includes training the machine learning model based on one or more losses associated with the first output frame and the first target frame.
A system for performing digital rights protected content playing includes a computing platform having a hardware processor and a memory storing a software code. The hardware processor executes the software code to receive, from a first user, metadata identifying one or more content segment(s), determine whether the first user has a right to access the content segment(s), produce a playlist using the metadata when the first user has the right, and output the playlist to the first user and/or a second user. The hardware processor may further execute the software code to receive the playlist from the second user, determine each content segment identified by the playlist that the second user has a right to access, generate a content compilation that omits any content segment identified by the playlist to which the second user lacks the right, and playout the content compilation to the second user.
A system includes a computing platform having processing hardware, and a memory storing software code. The processing hardware is configured to execute the software code to receive an image having a plurality of image regions, determine a boundary of each of the image regions to identify a plurality of bounded image regions, and identify, within each of the bounded image regions, one or more image sub-regions to identify a plurality of image sub-regions. The processing hardware is further configured to execute the software code to identify, within each of the bounded image regions, one or more first features, respectively, identify, within each of the image sub-regions, one or more second features, respectively, and provided an annotated image by annotating each of the bounded image regions using the respective first features and annotating each of the image sub-regions using the respective second features.
G06V 10/25 - Determination of region of interest [ROI] or a volume of interest [VOI]
G06V 10/46 - Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
G06V 20/70 - Labelling scene content, e.g. deriving syntactic or semantic representations
A system includes a computing platform having processing hardware and a memory storing a software code. The processing hardware is configured to execute the software code to receive input data from a user, determine, using the input data, an intent of the user and a commentator persona for providing a commentary to the user, and obtain, based on the input data, content data for use in the commentary. The processing hardware is further configured to execute the software code to generate, based on the intent of the user and using the content data, a script for the commentary, transform the script, using the commentator persona, to a commentator-specific script for the commentary, and output the commentary to the user, using the commentator-specific script.
ETH Zürich (Eidgenössische Technische Hochschule Zürich) (Switzerland)
Inventor
Winberg, Sebastian
Chandran, Prashanth
Gotardo, Paulo Fabiano Urnau
Zoss, Gaspard
Bradley, Derek Edward
Abstract
Embodiments of the present disclosure are directed to methods and systems for generating three-dimensional (3D) models and facial hair models representative of subjects (e.g., actors or actresses) using facial scanning technology. Methods accord to embodiments may be useful for performing facial capture on subjects with dense facial hair. Initial subject facial data, including facial frames and facial performance frames (e.g., images of the subject collected from a capture system) can be used to accurately predict the structure of the subject’s face underneath their facial hair to produce a reference 3D facial shape of the subject. Likewise, image processing techniques can be used to identify facial hairs and generate a reference facial hair model. The reference 3D facial shape and reference facial hair mode can subsequently be used to generate performance 3D facial shapes and a performance facial hair model corresponding to a performance by the subject (e.g., reciting dialog).
Certain aspects of the present disclosure provide techniques for automated updates to code deployment pipelines. This includes identifying a proposed change to a plurality of source code repositories based on parsing a change template. It further includes determining one or more changes to one or more files in each respective source code repository, of the plurality of source code repositories, based on the proposed change. It further includes changing the one or more files in each respective source code repository, based on the determined one or more changes, and building a project in each respective source code repository using the changed one or more files.
A system for performing synergistic object tracking and pattern recognition for event representation includes a computing platform having processing hardware and a system memory storing a software code. The processing hardware is configured to execute the software code to receive event data corresponding to one or more propertie(s) of an object, to generate, using the event data, a location data estimating a location of each of multiple predetermined landmarks of the object, and to predict, using one or both of the event data and the location data, a pattern corresponding to the propertie(s) of the object. The processing hardware is further configured to execute the software code to update, using the predicted pattern, the location data, and to merge the updated location data and the predicted pattern to provide merged data.
A system includes a computing platform including processing hardware and a system memory storing software code. The processing hardware executes the software code to receive a transcript of speech by a user, generate a phoneme stream corresponding to the transcript, partition the phoneme stream into words, and aggregate subsets of the words to form candidate sentences. The software code is further executed to determine, using one or both of an entity identified in the transcript and a history of the user, one or both of a user intent and a context of the speech by the user, rank the candidate sentences and the transcript based on one or both of the user intent and the context of the speech by the user, and identify, based on the ranking, one of the candidate sentences or the transcript as the best transcription of the speech by the user.
ETH Zürich (Eidgenössische Technische Hochschule Zürich) (Switzerland)
Inventor
Bradley, Derek Edward
Chandran, Prashanth
Urnau Gotardo, Paulo Fabiano
Otto, Christopher Andreas
Serifi, Agon
Zoss, Gaspard
Abstract
One embodiment of the present invention sets forth a technique for performing shape and appearance reconstruction. The technique includes generating a first set of renderings associated with an object based on a set of parameters that represent a reconstruction of the object in a first target image. The technique also includes producing, via a neural network, a first set of corrections associated with at least a portion of the set of parameters based on the first target image and the first set of renderings. The technique further includes generating an updated reconstruction of the object based on the first set of corrections.
A system for performing automated multi-persona response generation includes processing hardware, a display, and a memory storing a software code. The processing hardware executes the software code to receive input data describing an action and identifying a multiple interaction profiles corresponding respectively to multiple participants in the action, obtain the interaction profiles, and simulate execution of the action with respect to each of the participants. The processing hardware is further configured to execute the software code to generate, using the interaction profiles, a respective response to the action for each of the participants to provide multiple responses. In various implementations, one or more of those multiple responses may be used to train additional artificial intelligence (AI) systems, or may be rendered to an output device in the form of one or more of a display, an audio output device, or a robot, for example.
A conductor connector guard adapted for enclosing or housing one-to-many junctions between separable connectors. The guard is configured to house separable connectors for 3-channel, 5-channel, or 7-channel power distribution, with some of the embodiments presented being configured to house five pairs of separable connectors, with the connectors taking the form of single-pole conductor connectors in some cases. The guard is designed to comply with safety requirements for power distribution junctions that may be positioned within a public space. The guard is configured to provide a compact solution when contrasted with prior configurations involving physical barriers around power distribution junctions or involving physical objects such as trash cannisters and planters being placed over junctions. The guard is also adapted to provide enhanced public safety for spliced or joined conductors and to sequester energy sources in public spaces such as with a lock and tag out-type device.
According to one implementation, a system includes a computing platform having processing hardware, and a system memory storing a software code. The processing hardware is configured to execute the software code to receive a vocabulary, identify words from the vocabulary for use in extending the vocabulary, pair each of those words with every other of those words to provide word pairs, and output the word pairs to a vocabulary administrator. The software code also receives word pair characterizations identifying each of the word pairs as one of similar, dissimilar, or neither similar nor dissimilar, configures, based on the word pair characterizations, a multi-dimensional vector space including multiple embedding vectors each corresponding respectively to one of the identified words, and cross-references each of those words with its corresponding embedding vector to produce an extended vocabulary corresponding to the received vocabulary.
ETH Zürich (Eidgenössische Technische Hochschule Zürich) (Switzerland)
Inventor
Zoss, Gaspard
Gözcü, Baran
Solenthaler, Barbara
Yang, Lingchen
Kim, Byungsoo
Abstract
One embodiment of the present invention sets forth a technique for generating actuation values based on a target shape such that the actuation values cause a simulator to output a simulated soft body that matches the target shape. The technique includes inputting a latent code that represents a target shape and a point on a geometric mesh into a first machine learning model. The technique further includes generating, via execution of the first machine learning model, one or more simulator control values that specify a deformation of the geometric mesh, where each of the simulator control values is based on the latent code and corresponds to the input point, and generating, via execution of the simulator, a simulated soft body based on the one or more simulator control values and the geometric mesh. The technique further includes causing the simulated soft body to be outputted to a computing device.
Techniques for buffering data over high bandwidth networks are provided. A first portion of data is downloaded, by a device, into a buffer at a first download speed via a first network connection. Upon determining that the device is downloading data via a second network connection, at a second download speed greater than the first download speed, a second portion of data is downloaded, via the second network connection, into a cache.
A system for performing user-initiated content modification includes a computing platform having processing hardware and a system memory storing a software code. The processing hardware is configured to execute the software code to receive a request to perform a modification to content, determine, in response to the request, whether the modification is permissible or impermissible, and when the modification is determined to be impermissible, deny the request. When the modification is determined to be permissible, the processing hardware is configured to further execute the software code to obtain the content, obtain or produce alternate content for use in modifying the content per the request, and perform the modification to the content, using the alternate content, to provide modified content.
H04N 21/2668 - Creating a channel for a dedicated end-user group, e.g. by inserting targeted commercials into a video stream based on end-user profiles
H04N 21/239 - Interfacing the upstream path of the transmission network, e.g. prioritizing client requests
H04N 21/25 - Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication or learning user preferences for recommending movies
90.
Facial animation retargeting using a patch blend-shape solver
ETH Zürich (Eidgenössische Technische Hochschule Zürich) (Switzerland)
Inventor
Chandran, Prashanth
Ciccone, Loïc Florian
Bradley, Derek Edward
Abstract
Methods and systems for performing facial retargeting using a patch-based technique are disclosed. One or more three-dimensional (3D) representations of a source character's (e.g., a human actor's) face can be transferred onto one or more corresponding representations of a target character's (e.g., a cartoon character's) face, enabling filmmakers to transfer a performance by a source character to a target character. The source character's 3D facial shape can separated into patches. For each patch, a patch combination (representing that patch as a combination of source reference patches) can be determined. The patch combinations and target reference patches can then be used to create target patches corresponding to the target character. The target patches can be combined using an anatomical local model solver to produce a 3D facial shape corresponding to the target character, effectively transferring a facial performance by the source character to the target character.
ETH Zürich (Eidgenössische Technische Hochschule Zürich) (Switzerland)
Inventor
Bradley, Derek Edward
Gotardo, Paulo Fabiano Urnau
Zoss, Gaspard
Chandran, Prashanth
Winberg, Sebastian
Abstract
Embodiments of the present disclosure are directed to methods and systems for generating three-dimensional (3D) models and facial hair models representative of subjects (e.g., actors or actresses) using facial scanning technology. Methods accord to embodiments may be useful for performing facial capture on subjects with dense facial hair. Initial subject facial data, including facial frames and facial performance frames (e.g., images of the subject collected from a capture system) can be used to accurately predict the structure of the subject's face underneath their facial hair to produce a reference 3D facial shape of the subject. Likewise, image processing techniques can be used to identify facial hairs and generate a reference facial hair model. The reference 3D facial shape and reference facial hair mode can subsequently be used to generate performance 3D facial shapes and a performance facial hair model corresponding to a performance by the subject (e.g., reciting dialog).
Embodiments herein describe a CDN where anycast routing is used to identify a load balancer for selecting a cache in the CDN to use to deliver a requested object to a user. In one embodiment, the user performs a DNS lookup to identify an anycast IP address for a plurality of load balancers in the CDN. The user can then initiate anycast routing using the anycast IP address to automatically identify the closest load balancer. Once the identified balancer selects the cache, the load balancer can close the anycast connection with the user device and use an HTTP redirect to provide the user device with a unicast path to the selected cache. The user device can then establish a unicast connection with the cache to retrieve (e.g., stream) the object.
H04L 67/1036 - Load balancing of requests to servers for services different from user content provisioning, e.g. load balancing across domain name servers
H04L 67/02 - Protocols based on web technology, e.g. hypertext transfer protocol [HTTP]
H04L 67/101 - Server selection for load balancing based on network conditions
H04L 61/4511 - Network directories; Name-to-address mapping using standardised directory access protocols using domain name system [DNS]
H04L 67/568 - Storing data temporarily at an intermediate stage, e.g. caching
H04L 67/10 - Protocols in which an application is distributed across nodes in the network
H04L 67/1004 - Server selection for load balancing
H04L 67/1021 - Server selection for load balancing based on client or server locations
Systems and methods generating a haptic output response are disclosed. Video content is displayed on a display. A location of a user touch on the display is detected while the video content is being displayed. A region of interest in the video content is determined based on the location of the user touch. And a haptic output response is generated to a user. A characteristic of the haptic output response is determined using one or more characteristics of the region of interest.
G09B 21/00 - Teaching, or communicating with, the blind, deaf or mute
G06F 3/041 - Digitisers, e.g. for touch screens or touch pads, characterised by the transducing means
G06V 10/46 - Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
G06V 20/40 - Scenes; Scene-specific elements in video content
A system for granting access to an account at an access device includes a computer server having a hardware processor and a memory storing a software code. The hardware processor executes the software code to receive a login request from the access device through a first communications socket, open a second communications socket between the access device and the computer server, transmit a verification request message including a required call-to-action to a verification device through a third communications socket, and receive a verification response message verifying that the required call-to-action has been completed at the verification device. Upon receiving the verification response message, the software code sends an access token for accessing the account to the access device through the second communications socket, receives the access token from the access device, and grants the access device access to the account.
According to one implementation, a method of guiding an interaction between a companion module and a user includes identifying a media content for playout by a media player device, transmitting, to a remote server, an interaction schema request identifying the media content and the companion module, and receiving, from the remote server, behavioral manifest data including an instruction for guiding the interaction between the companion module and the user. Such a method also includes obtaining a play head state of the media player device and a playout timecode state of the media content, and identifying a user interaction behavior for the companion module based on the behavioral manifest data, the play head state of the media player device, and the playout timecode state of the media content.
H04N 21/43 - Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronizing decoder's clock; Client middleware
H04N 21/8547 - Content authoring involving timestamps for synchronizing content
A method to project content from a moving projector. The method includes analyzing an object to identify a projection surface; determining a first position of a projector relative to the projection surface; modifying a first frame of a content for projection onto the projection surface based on the projection surface and the first position of the projector; projecting the first frame of the content from the projector onto the projection surface; determining a second position of the projector relative to the projection surface; modifying a second frame of the content based on the projection surface and the second position; and projecting the second frame of the content from the projector onto the projection surface.
Techniques are disclosed for characterizing audience engagement with one or more characters in a media content item. In some embodiments, an audience engagement characterization application processes sensor data; such as video data capturing the faces of one or more audience members consuming a media content item, to generate an audience emotion signal. The characterization application also processes the media content item to generate a character emotion signal associated with one or more characters in the media content item. Then, the characterization application determines an audience engagement score based on an amount of alignment and/or misalignment between the audience emotion signal and the character emotion signal.
H04N 21/442 - Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed or the storage space available from the internal hard disk
H04N 21/466 - Learning process for intelligent management, e.g. learning user preferences for recommending movies
98.
TECHNIQUES FOR IMPROVED LIGHTING MODELS FOR APPEARANCE CAPTURE
ETH Zürich (Eidgenössische Technische Hochschule Zürich) (Switzerland)
Inventor
Urnau Gotardo, Paulo Fabiano
Bradley, Derek Edward
Zoss, Gaspard
Riviere, Jeremy
Chandran, Prashanth
Xu, Yingyan
Abstract
Various embodiments include a system for rendering an object, such as human skin or a human head, from captured appearance data comprising a plurality of texels. The system includes a processor executing a texture space indirect illumination module. The system determines texture coordinates of a vector originating from a first texel where the vector intersects a second texel. The system renders the second texel from the viewpoint of the first texel based on appearance data at the second texel. Based on the rendering of the second texel, the system determines an indirect lighting intensity incident to the first texel from the second texel. The system updates appearance data at the first texel based on a direct lighting intensity and the indirect lighting intensity. The system renders the first texel based on the updated appearance data at the first texel.
According to one exemplary implementation, a system includes a hardware processor and a system memory storing a software code. The hardware processor is configured to execute the software code to receive, from a user device, a request for a non-fungible token (NFT) based on the presence of a user of the user device in a venue, receive sensor data identifying a location of the user device, and obtain camera data from the venue, the camera data depicting at least one of the user of the user device or a field of view of the user relative to the venue. The hardware processor is further configured to execute the software code to mint the NFT, using the sensor data and the camera data, wherein the NFT depicts at least one of a portion of an object situated within the venue or an event occurring at the venue.
ETH Zürich (Eidgenössische Technische Hochschule Zürich) (Switzerland)
Inventor
Urnau Gotardo, Paulo Fabiano
Bradley, Derek Edward
Zoss, Gaspard
Riviere, Jeremy
Chandran, Prashanth
Xu, Yingyan
Abstract
Various embodiments include a system for rendering an object, such as human skin or a human head, from captured appearance data. The system includes a processor executing a near field lighting reconstruction module. The system determines at least one of a three-dimensional (3D) position or a 3D orientation of a lighting unit based on a plurality of captured images of a mirror sphere. For each point light source in a plurality of point light sources included in the lighting unit, the system determines an intensity associated with the point light source. The system determines captures appearance data of the object, where the object is illuminated by the lighting unit. The system renders an image of the object based on the appearance data and the intensities associated with each point light source in the plurality of point light sources.