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.

Abstract: A gameplay system for providing interactivity based on a piece of merchandise or another object held or worn by a player. The system includes an object identification assembly, which may be a modified computer vision system, that operates to sense, for a gameplay participant, the presence and identity of a particular augmenting object such as themed merchandise worn or held by a rider of a theme park ride. The sensing may be performed by detecting an on-and-off pattern of light emitted from the object. In response, the gameplay system selects or modifies the interactive features of gameplay associated with the participant based on the identified augmenting object. For example, the interactive features are selected from memory by the gameplay system controller to match a character associated with the augmenting object during any interactive gameplay. This allows the gameplay participant to choose their abilities or powers for gameplay.

Abstract: According to one implementation, an image enhancement system includes a computing platform including a hardware processor and a system memory storing a software code configured to provide a normalizing flow based generative model trained using an objective function. The hardware processor executes the software code to receive an input image, transform the input image to a latent space representation of the input image using the normalizing flow based generative model, and perform an optimization of the latent space representation of the input image to identify an enhanced latent space representation of the input image. The software code then uses the normalizing flow based generative model to reverse transform the enhanced latent space representation of the input image to an enhanced image corresponding to the input image.

Abstract: Systems and methods are presented for immersive and simultaneous animation in a mixed reality environment. Techniques disclosed represent a physical object, present at a scene, in a 3D space of a virtual environment associated with the scene. A virtual element is posed relative to the representation of the physical object in the virtual environment. The virtual element is displayed to users from a perspective of each user in the virtual environment. Responsive to an interaction of one user with the virtual element, an edit command is generated and the pose of the virtual element is adjusted in the virtual environment according to the edit command. The display of the virtual element to the users is then updated according to the adjusted pose. When simultaneous and conflicting edit commands are generated by collaborating users, policies to reconcile the conflicting edit commands are disclosed.

Abstract: Some implementations of the disclosure are directed to capturing facial training data for one or more subjects, the captured facial training data including each of the one or more subject's facial skin geometry tracked over a plurality of times and the subject's corresponding jaw poses for each of those plurality of times; and using the captured facial training data to create a model that provides a mapping from skin motion to jaw motion. Additional implementations of the disclosure are directed to determining a facial skin geometry of a subject; using a model that provides a mapping from skin motion to jaw motion to predict a motion of the subject's jaw from a rest pose given the facial skin geometry; and determining a jaw pose of the subject using the predicted motion of the subject's jaw.

Abstract: A system adapted for stopping and launching boats. The system includes a boat, in moving water, with an interaction member extending outward from its hull. The system includes a boat capture and launch assembly that includes: (a) a body; (b) a drive mechanism rotating the body; (c) a capture arm with a first contact surface on the body; and (d) a launch arm with a second contact surface on the body. When the body is rotated into a first position, the capture arm extends outward into the water to catch the interaction member and stop the boat. When the body is rotated through a range of second positions, the interaction member abuts the second contact surface and a launch force is applied to accelerate the boat in the direction of travel of the boat and water.

Abstract: The exemplary embodiments relate to converting standard dynamic range (SDR) content to high dynamic range (HDR) content. An SDR image may be decomposed into a base layer of the SDR image that includes low frequency information from the SDR image and a detail layer of the SDR image that includes high frequency information from the SDR image. A base layer of an HDR image may be generated using the base layer of the SDR image and a detail layer of the HDR image may be generated using the detail layer of the SDR image. An HDR image is then generated using the base layer of the HDR image and the detail layer of the HDR image.

Abstract: Software defined network orchestration to manage media flows for broadcast with public cloud networks is provided by identifying a media flow at a media production facility for multicast transmission; registering the media flow to a registration database; migrating the media flow from multicast transmission to unicast transmission; transmitting the media flow to a public cloud network facility; and updating the registration database with a location of the media flow in the public cloud network facility. Once registered, a media flow management system allows any authorized device to request for a media flow; and in response locates the media flow based on a registration database indicating a location of the media flow (whether in the public cloud network facility, on a common carrier, or in a production facility); receives access to the media flow at the location; and allows the authorized device to consume the media flow.

Abstract: Techniques and systems for immersive interactions are provided. A first smart device available for interaction is identified by an interactivity system. A first physical game piece is detected by the interactivity system, and a first identifier of the first physical game piece is received, from the first physical game piece, via one or more wireless signals. A first set of predefined effects that can be performed using the first smart device is identified based on the first identifier. The first smart device is instructed to perform at least a first predefined effect of the first set of predefined effects.

Abstract: A system for synchronizing media content playout includes a computing platform having a hardware processor and a system memory storing a software code. The hardware processor executes the software code to receive a first state message from a first media player playing a first media content and a second state message from a second media player playing a second media content, the first media content and the second media content being the same media content. The software code further determines a coordination state for playout of the first media content and the second media content based on one or more of the first and second state messages, and transmits a first coordination message including the coordination state to the first media player and a second coordination message including the coordination state to the second media player to synchronize playout of the first media content and the second media content.

Abstract: A system and method for public API authentication by an API server includes receiving from a client/app a PK request for a Partial Key (PK), having a User ID, Session ID, rolling hash function (Fn2) version defining a client/app hash function (Fn2), and a received Temporary Key (TK); validating the received TK using Fn2 with the Session ID and either an Initial Key (IK) or a current PK; when the TK validation is complete, sending a PK calculated using a PK hash function (Fn1) with the User ID and a slot-generated rolling random number; receiving an API request for an API service having the User ID, Session ID, Fn2 version, and a received Authorization Key (AK); validating the received AK using Fn2 with the Session ID and the current PK; and when AK validation is complete, sending a successful response from the API service.

Abstract: In one embodiment, a character engine models a character that interacts with users. The character engine receives user input data from a user device, and analyzes the user input data to determine a user intent and an assessment domain. Subsequently, the character engine selects inference algorithm(s) that include machine learning capabilities based on the intent and the assessment domain. The character engine computes a response to the user input data based on the selected inference algorithm(s) and a set of personality characteristics that are associated with the character. Finally, the character engine causes the user device to output the response to the user. In this fashion, the character engine includes sensing functionality, thinking and learning functionality, and expressing functionality.

Abstract: Non-linear management of real time sequential data in cloud instances via time constraints is provided by: receiving, at a cloud network facility from a media production facility, a selection of component media flows for assembly into an assembled media package; identifying a time constraint for assembling component media flows into the assembled media package; identifying processes to perform with respect to the component media flows during assembly of the assembled media package; in response to an estimated time to perform the processes with respect to the component media flows exceeding the time constraint, adjusting performance of the processes with respect to the component media flows to increase a speed of performance; and performing, in the cloud network facility, the processes with respect to the component media flows as adjusted to produce the assembled media package.

Abstract: A recommendation engine generates contextually relevant media programs tailored to the specific context of the user. The recommendation engine interfaces with one or more Internet-of-Things (IoT) appliances to determine one or more food items stored within the IoT appliance(s) and/or one or more cooking capabilities associated with the IoT appliance(s). The recommendation engine also gathers profile data that reflects a set of preferences associated with the user. Based on the gathered data, the recommendation engine queries a database of cooking program data associated with various cooking programs. The recommendation engine obtains cooking program data that is contextually relevant to the user and then generates a contextual cooking program that describes how to perform a given recipe. During playback of the contextual cooking program, the recommendation engine can modify the contextual cooking program based on the progress of the user in following the recipe.

Abstract: A method of monitoring a user performing a handwashing procedure, including: recognizing a user based upon biometric information obtained at a handwashing station, a user; detecting use of a component of the handwashing station; capturing image data of the user at the handwashing station; capturing data describing components and consumables used at the handwashing station; analyzing, using a processor, the image data; determining compliance with a handwashing procedure based at least in part on the analysis of the image data and the use of the component; and communicating compliance information.

Abstract: Various embodiments set forth a computer-implemented method for selectively caching digital content for streaming, comprising analyzing a set of traffic data associated with streaming of a plurality of media items in a first time period, selecting, based on the set of traffic data, a first media item for storage in a cache during a second time period, and storing in the cache, prior to receiving a request for the first media item in the second time period, at least a portion of a first media stream of a plurality of media streams associated with the first media item, and at least a portion of a second media stream of the plurality of media streams, wherein each media stream in the plurality of media streams is encoded with different characteristics.

Abstract: A robot including a hybrid fan-based and fluid-based propulsion system to provide thrust, such as deceleration during fall to create a smooth landing or to provide a quick reduction in velocity, and to provide actuation/controlled motion, such as to hover after quick deceleration and to control orientation or pose. The hybrid propulsion system uses discharging of pressurized fluid and exhausted gas (or fluid in some cases) from ducted fans (or propellers, impellers, and the like) to provide controlled thrust and/or lift forces. The hybrid propulsion system uses of pressurized fluid for generating larger or primary thrust and quick changes in velocity. The hybrid propulsion system includes a fan-based propulsion assembly with ducted fans that use environmental air (or fluids) to provide lower or secondary thrust. Both types of propulsion can be integrated into a robot or robotic figure to move the robot during flight (e.g., during falling or hovering).

Abstract: A storytelling system includes location-based content (LBC) servers and LBC effects databases for use in a real-world story venue, and a computing platform communicatively coupled to those LBC servers and databases. A processor of the computing platform executes software code to receive story selection data from a user device, obtain a story template including story arcs each associated with one or more LBC effect(s) that corresponds to the story selection data, and determine, using a location of the user device, one of the story arcs as an active story arc. The software code also identifies an LBC interaction zone for the LBC effect(s), designates one LBC server and one LBC effects database for supporting the active story arc, and distributes one or more of the LBC effect(s) to the designated database. The designated server enables instantiation of the one or more of the LBC effect(s) at the LBC interaction zone.

Abstract: The embodiments herein describe a CDN that uses revalidation to force edge servers to refresh their cached objects (i.e., download new copies of the objects from the origin data center). Revalidation can be used by edge servers to determine whether an object currently cached at the edge servers has been updated or changed in the origin data center. The embodiments herein leverage revalidation to perform a forced refresh to force the edge servers to refresh their cached object, regardless whether those cached objects match the objects stored in the origin data center. A forced refresh can be used when there is a network connection that may have caused the cached objects to become corrupted. Moreover, the forced refresh can be performed in batches so as not to overwhelm the network and computing resources in the CDN.