Archive year
2026
164 Disney / Lucasfilm filings published in 2026 and read by the sweep: 84 kept as hardware or notable software, 80 set aside. Kept filings are ranked by kinship to saber, droid, animatronic and show-effect work.
High kinship (29)
- Automated parallelized trajectory analysis
A physics-simulation method that predicts where an object released from a moving robotic device will land, timing the release across each step of its motion -- handy for tuning a droid's throw or toss gag.
- Pose-aware neural inverse kinematics
Uses a neural network to solve a virtual character's joint poses from constraints — inverse-kinematics tech relevant to robotics/animatronic motion control.
- DISPLAY SYSTEM FOR PRODUCING WIDE-ANGLE, RETROREFLECTIVE REAL IMAGES
A ring of beamsplitter/retroreflector stations that projects a real, glasses-free 3D image floating in space that a whole crowd can walk around and view from any angle — a scalable Pepper's-ghost-style theme-park effect.
- THEMED AERIAL VEHICLE ENTERTAINMENT PLATFORM FOR PROVIDING DYNAMICALLY-COORDINATED SHOW
Show drones wrapped in themed shells (e.g. a spacecraft or flying character) fly dynamically-coordinated routines with onboard lights instead of a fixed flight plan — direct animatronics/show-tech interest.
- AUTONOMOUS HUMAN-ROBOT INTERACTION VIA DIFFUSION-BASED OPERATOR IMITATION
Trains a diffusion model so a robot can autonomously read and respond to a human's actions — the control layer for a free-roaming character droid.
- Stable pose estimation with analysis by synthesis
Analysis-by-synthesis pose estimation that stays stable frame to frame — tracking tech for guests, performers, or robots.
- Environmental factor-based surface maintenance system
Autonomous surface-maintenance robot that adapts its cleaning to detected environmental factors.
- Versatile motion priors for artist-directable physical characters
Motion priors that let artists direct a physical (robotic) character's movement while keeping it physically plausible — Bächer's bipedal-droid line.
- Ride system featuring a free-range vehicle platform
Ride system with a free-range vehicle platform driving over uneven terrain rather than a fixed track path.
- OPPORTUNISTIC PROJECTION ONTO DYNAMIC LIGHT REFLECTIVE MATERIAL
Camera-tracked projection onto light-reflective material that moves through a space (costumes, props, set pieces) — Smoot.
- METHODS FOR ALIGNMENT OF PLAN TO EXECUTION USING MULTIMODAL CUES FOR SCRIPT-BASED PERFORMANCE MONITORING
Aligns a scripted performance plan to its live execution using multimodal cues — show-control for animatronic/robot performances.
- SKIN EMBEDDED MARKERS FOR DYNAMIC PROJECTION MAPPING OF ANIMATRONIC FIGURES
A system includes an animatronic skin having one or more embedded markers configured to provide a light signature, a camera configured to detect the light signature, a projector configured to project content onto the an…
- HYBRID APPROACH TO PERFORM DYNAMIC PROJECTION MAPPING ONTO ROBOTIC FIGURES IN REAL-TIME
Systems, methods and apparatuses for a hybrid approach to perform dynamic projection mapping onto robots in real-time are discussed herein.
- HYBRID APPROACH TO PERFORM DYNAMIC PROJECTION MAPPING ONTO ROBOTIC FIGURES IN REAL-TIME
Systems, methods and apparatuses for a hybrid approach to perform dynamic projection mapping onto robotic figures in real-time are discussed herein.
- DYNAMIC MECHANICAL SKIN STRUCTURES FOR ANIMATRONIC FIGURES
An animatronic may include a shell, a skin coupled to the shell, and an actuator configured to selectively create a deformation of the skin, wherein the deformation is configured to complement content projected onto the…
- PROJECTION MAPPING USING DYNAMIC BLEND MAPS
A computer-implemented technique for projecting light includes generating, based on a pose of an object within the scene, projector calibration data associated with one or more projectors, and scene geometry data associ…
- DATA-DRIVEN CONTENT INTERACTION
Interprets user input across LED, haptic and wearable channels to drive content — new inventor cluster, worth a look.
- DYNAMIC WEIGHT ADJUSTMENT FOR MOTION CONTROL VIA MULTI-OBJECTIVE REINFORCEMENT LEARNING
One embodiment of the present invention sets forth a technique for controlling motion in an articulated object.
- ADAPTIVE MOTION CONTROL VIA MULTI-OBJECTIVE REINFORCEMENT LEARNING
One embodiment of the present invention sets forth a technique for controlling motion in an articulated object.
- System for generating a 360 degree viewable image on laminar fog screen
A system may comprise a fog generating unit configured to generate fog; a fog distributing unit comprising: an intake component configured to receive the fog generated by the fog generating unit, a fog manifold configur…
- TECHNIQUES FOR CONTROLLING ACTUATED SYSTEMS USING NON-DETERMINISTIC DATA
A method for controlling an actuated system includes receiving one or more data points, where each data point included in the one or more data points associates animation data with a different point in time, computing a…
- METHODS AND SYSTEMS FOR ROBOT KEYFRAMING AND LEARNING LOCOMOTION WITH HIGH-LEVEL OBJECTIVES
Methods and systems for animated figure keyframing or physical robot keyframing and learning locomotion with high-level objectives are discussed herein.
- ATTENTION-BASED MAP ENCODING FOR GENERALIZED ROBOT LOCOMOTION
The present invention sets forth techniques for generating robot action commands directed to robot locomotion.
- AUTOMATED PARALLELIZED TRAJECTORY ANALYSIS
A method for determining a trajectory of an object includes: determining, via a processing element, the object of the moving device; generating, via the processing element, a physics model of a moving device including t…
- Wireless Power Delivery in a Fluid Environment
A system includes a reservoir containing a fluid, one or more objects in physical contact with the fluid, each of the one or more objects having a respective induction coil embedded therein, at least one conductive coil…
- Projection System Including a 3-D Extendable Projection Surface
A system includes a computing platform having a hardware processor and a system memory storing software code, as well as a projection unit controlled by the computing platform, the projection unit including a three-dime…
- BUBBLE PRODUCING TOY
A toy may include a wand defining a handle for gripping by a user to hold the toy, and a bubble generator configured to produce bubbles or foam.
- Server rack mountable automated test platform for transaction device testing
An automated test platform configured for use in testing and validating point of sale (POS) software and hardware.
- Display system for producing wide-angle, retroreflective real images
A wide-angle, retroreflective real image display system that produces real images that can be viewed and interacted with by a group of people who can move about a space to view the real image.
Some kinship (22)
- Modular omnidirectional actuated floors providing a reconfigurable video production environment
A stage floor of independently-moving tiles that push back against performers to simulate motion for the camera — an actuated show-floor rig for motion-platform builders.
- Selective visualization of an image using second pass-through of polarized light
A sign with two polarizing filters and a perforated image layer shows a different picture depending on which polarizer the viewer looks through -- a hidden-image effect useful for props or displays.
- SYSTEM FOR GENERATING COLLIMATED NON-COHERENT BRIGHT LIGHT USING A PLURALITY OF OPTICAL ELEMENTS
Combines two laser arrays with a phosphor converter to produce bright, non-coherent collimated light — projector/show-lighting hardware worth watching for prop and stage applications.
- RIDER INCLUSIVE TRANSPORT AIDE
A self-leveling platform on a theme-park ride vehicle that stays flat (for wheelchairs/gear) as the vehicle tilts over water or uneven ground.
- TRAINING OF MACHINE LEARNING MODELS WITH HARDWARE-IN-THE-LOOP SIMULATIONS
Trains fault-tolerant control-system models by injecting simulated faults into hardware-in-the-loop sensor emulations — relevant to robotics/animatronic control tuning.
- Water harp with optical sensing of the break of flowing water streams
A themed 'water harp' show prop: light beamed through falling water streams plays a musical note when a hand breaks the beam — classic interactive show-tech hardware.
- Memories for virtual characters
Gives a virtual character persistent memories of past interactions so later encounters with a guest reference them.
- INTERACTIVE IMMERSIVE EXPERIENCES USING MACHINE LEARNING
Machine learning that tunes an interactive immersive experience to how guests are engaging with it.
- HIGH-RESOLUTION VIRTUAL VEHICLE PRESENCE SENSOR
High-resolution virtual presence sensor that tells a ride system exactly where a vehicle is without physical trip sensors.
- ELEVATION CHANGE ILLUSION THROUGH SET DESIGN
Set-design trick that makes a track section feel like a bigger elevation change than it is.
- SYSTEM AND METHOD FOR GENERATING AN IMAGE USING A NATURAL PROJECTION MEDIUM
A system includes a light source configured to emit light; a lens assembly including an optical component and at least one lens; a fiber optic cable configured to transmit the light emitted from the light source to the …
- VISUALLY CONSISTENT MULTI-VIEW NOVEL VIEW SYNTHESIS WITH VIDEO DIFFUSION
Methods and systems for performing novel view synthesis in addition to various image-to-image diffusion processes using machine learning models are disclosed.
- METHODS AND SYSTEMS FOR PERFORMING NOVEL-VIEW SYNTHESIS
Methods and systems for performing novel view synthesis in addition to various image-to-image diffusion processes using machine learning models are disclosed.
- METHOD AND SYSTEM FOR GENERATING TRAINING DATA FOR EFFICIENT AND ACCURATE TRAINING OF NOVEL VIEW SYNTHESIS MODELS
Methods and systems for performing novel view synthesis in addition to various image-to-image diffusion processes using machine learning models are disclosed.
- PATH GUIDING USING NEURAL RADIANCE CACHING WITH DEFENSIVE RESAMPLED IMPORTANCE SAMPLING AND ADAPTIVE CANDIDATE ALLOCATION
The present disclosure relates to computer-implemented path guiding in a rendering system.
- System for generating collimated non-coherent bright light using a plurality of optical elements
A system, comprising: a first array of lasers; a first linear polarizer; a second array of lasers; a second linear polarizer provided at an angle with respect to the first linear polarizer; a first beamsplitter coupled …
- COMPLIANT MECHANISM SUSPENSION SYSTEM
A suspension system may include a plate, a frame extending at least partially around the plate, and a resilient member extending between to connect the plate to the frame.
- CONTENT PROJECTION SYSTEM FOR PROJECTING CONTENT ONTO RIDE VEHICLES
A content projection system may include a projector configured to project content onto a first ride vehicle, a tracking system configured to track (e.g., a position and an orientation of) the first ride vehicle within a…
- Clearance envelope management for regulating vehicle movement
A method may include comparing, using a virtual map, a position of a ride vehicle to a position of an obstacle in an operating environment; and querying a virtual map of an operating environment, the virtual map includi…
- Training of machine learning models with hardware-in-the-loop simulations
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 se…
- Accelerated sum-of-squares collision detection for time-varying curved shapes
A method for detecting collisions associated with a simulation includes generating a plurality of spline-based representations associated with a plurality of objects.
- PLUSH TOY WITH STRAP FOR SECURING TO ITEMS
In some implementations, a plush toy may comprise a strap receiving portion provided on a surface of the plush toy.
Distant (33)
- Emotionally Responsive Artificial Intelligence Interactive Character
An interactive AI character asks a guest to share a memory, scores how closely it matches memories it already holds, and shifts its emotional response when nothing matches well.
- AI Generated Creative Content Based on Shared Memories
The same AI-character memory system turned toward storytelling: it pulls a guest's shared memory and generates new creative content, like a song or tale, out of it.
- EVENT DETECTION BASED ON ANALYSIS OF VIDEO DATA
A vision system that watches video to spot a person and a specific event happening to them, then scores how confident it is before acting on the detection.
- Real-time animation motion capture
Records a performer's body and facial motion, then grafts in motion-capture data from a second performance to splice two actors' takes into one animated character.
- Anonymized Human Body Tracking
Uses a sensor to spot a person's body parts and track an anonymized, skeleton-like position in real time -- markerless motion tracking an animatronic or interactive droid could react to.
- MULTI-LAYER REPRESENTATIONS FOR ARTISTICALLY CONTROLLABLE STEREO CONVERSION
Uses machine learning to predict depth and split a 2D image into layers so artists can convert flat footage into stereo 3D with manual control over the depth effect.
- NEURAL VIEW-DEPENDENT PRINCIPLED BRDF ADAPTER
Trains a neural network to adapt one material's light-reflecting properties to mimic another from any viewing angle -- a shortcut for realistic material rendering in CG.
- Crowd removal and automatic focus for images
Splits a photo into segments, uses depth data to separate subjects from bystanders, then automatically removes the crowd and focuses on the subject -- useful for cleaning up theme-park photo captures.
- Interruption Response by an Artificial Intelligence Character
Gives a conversational AI character (think interactive droids/animatronics) the ability to notice when it's been interrupted mid-sentence and pick a smart way to respond instead of talking over people.
- BOOTSTRAPPED ROTOSCOPE-MATTING TRAINING
Trains a computer-vision model to auto-rotoscope/matte VFX footage by synthesizing composite training data from real and rendered foreground objects.
- Image-Space UV Map-Based Tracking of Image Regions in Video
Tracks a region on a moving object across video frames using AI-predicted UV maps so a mask can follow it automatically — a VFX/compositing tracking tool.
- AI generated creative content based on shared memories
An AI interactive character (AIIC) draws out a user's memories, then generates personalized creative content from them — Disney's memory-aware companion-bot tech.
- Emotionally responsive artificial intelligence interactive character
An AI interactive character (AIIC) reads emotional cues from what a user says and adjusts its responses to match — companion-bot emotional intelligence.
- User responsive augmented reality enhancement of moving images
AR headset that augments a moving image (film/show) in response to what the wearer does.
- Reducing domain shift in neural motion controllers
Trains a neural controller to translate a control signal into realistic character motion — locomotion-control ML relevant to animatronics/robot movement research.
- Providing assistance based on machine learning model analysis of video data
Computer-vision system that watches pool/water areas and flags a person in distress — real-time CV/tracking, relevant to anyone doing camera-based safety detection.
- NEURAL SHAPE DEFORMATION TRANSFER
One embodiment of the present invention sets forth a technique for generating a shape.
- TRIGGER FOR CAMERA TRAP
A system and method for capturing fames (e.g., images) in an environment.
- DATA-DRIVEN MODELING OF SECONDARY MOTION DYNAMICS USING GAUSSIAN SPLATTING
The present invention sets forth techniques for predicting motion in a 3D model.
- Identity-preserving image generation using diffusion models
One embodiment of the present invention sets forth a technique for performing identity-preserving image generation.
- BALL AND SOCKET MOUNT WITH PERMANENT BALL LOCKING
In some implementations, an assembly may comprise a ball comprising one or more first channels provided on an outer surface of the ball.
- Torque value adjustment system for use with click type torque wrenches to minimize repetitive motions
A system for adjusting torque value settings on click type torque wrenches.
- Interruption response by an artificial intelligence character
A system includes a hardware processor, a memory storing software code, and a machine learning (ML) model trained to detect an interruption to a conversation.
- PARAMETRIC LANDMARK DETECTION
One embodiment of the present invention sets forth a technique for performing landmark detection.
- PHYSICAL VARIANCE DETECTION AUTOMATA
The present invention sets forth a technique for performing automated physical variance detection.
- Graph simulation for facial micro features via pre-computed displacement texture maps
The present invention sets forth a technique for simulating wrinkles under dynamic facial expression.
- Depth-aware indexing of three-dimensional image depth bins
A system includes a hardware processor and a memory storing software code.
- Predicting a need for assistance by determining behavior based on machine learning model analysis of video data
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.
- FACTORIZED MOTION COMPLETION FOR PRECISE AND CHARACTER-AGNOSTIC MOTION DIFFUSION
The present invention sets forth techniques for animating a three-dimensional (3D) character model.
- Latency reduction for immersive content production systems
A method of content production may include receiving tracking information for a camera with a frustum configured to capture images of a subject in an immersive environment.
- TWO-DIMENSIONAL GAUSSIAN SPLATTING WITH TEXTURING FOR VIEW GENERATION IN COMPUTING SYSTEMS
Embodiments of the present disclosure provide generating one or more views of a scene from an input depicting the scene.
- SCREEN OUTPAINTING FOR IMMERSIVE EXPERIENCES
The present invention sets forth a technique for performing screen outpainting.
- Facial image swapping
Techniques for generating modified images using facial content information are disclosed.
Set aside (80)
Show the 80 filings the sweep read and did not keep
- MULTIMODAL AUDIO CHANNEL CLASSIFICATION
Set aside: streaming, codec
- IDENTITY-PRESERVING IMAGE GENERATION USING DIFFUSION MODELS
Set aside: dropped in weekly review
- MEDIA COMPONENT ALIGNMENT AND TRANSFER
Set aside: streaming, ad-tech, search/metadata
- Automatic detection of alignment between two audio signals
Set aside: dropped in weekly review
- TECHNIQUES FOR DENOISING VIDEOS
Set aside: dropped in weekly review
- Secure digital registration of contracts governing media content use
Set aside: search/metadata, security/drm, business-ops
- DYNAMIC PRODUCT DESIGN VIA GENERATIVE ARCHITECTURES
Set aside: recommendation
- AUDIO SYNCHRONIZATION THROUGH LARGE LANGUAGE MODELS AND DIFFERENTIAL EQUATION-BASED MEL SPECTOGRAM INVERSION
Set aside: dropped in weekly review
- DETECTING ANOMALOUS CONNECTIONS AND PROXY USAGE FOR NETWORK SESSIONS
Set aside: streaming, billing/fraud, security/drm
- SYSTEM AND METHOD TO REMOTELY CONTROL AND TEST MEDIA PLAYER DEVICES
Set aside: streaming, ad-tech
- System and method for monitoring procedure compliance
Set aside: dropped in weekly review
- Resilient content distribution network
Set aside: dropped in weekly review
- Machine learning model based condition and property detection
Set aside: dropped in weekly review
- DOMAIN SPECIFIC INTELLIGENCE USING NATURAL LANGUAGE
Set aside: nlp/localization, business-ops
- Machine learning model-based image noise synthesis
Set aside: dropped in weekly review
- TECHNIQUES FOR AUTOMATICALLY MONITORING ANIMAL BEHAVIORS AND GENERATING NOTIFICATIONS
Set aside: dropped in weekly review
- TECHNIQUES FOR DETECTING ANOMALIES IN AUDIO RECORDINGS FOR PREDICTIVE MAINTENANCE
Set aside: cpc prior 0.44
- OFFLINE REINFORCEMENT LEARNING FOR VIDEO ENCODER RATE CONTROL
Set aside: streaming, codec, cdn/infra
- Machine learning model-based acoustic signal-to-text translation
Set aside: dropped in weekly review
- Centrally managed remote storage encryption and decryption
Set aside: security/drm
- Temporally stable occlusion free caption rendering position in Z-plane for stereoscopic video
Set aside: dropped in weekly review
- ASSET STORAGE AND RETRIEVAL USING WORD PERMUTATIONS
Set aside: codec, business-ops
- MULTILINGUAL MULTIMEDIA TRANSLATION WITH SYNCHRONIZED AUDIO AND BODY GESTURES
Set aside: dropped in weekly review
- Deployable canopy system
Set aside: cpc prior 0.27
- Dynamic selection of candidate bitrates for video encoding
Set aside: streaming, codec
- INTELLIGENT CONTENT DELIVERY NETWORK (CDN) ENTITY ROUTING
Set aside: cdn/infra
- ARTIFACT PROCESSING IN VIDEO USING TEXTURE INFORMATION
Set aside: dropped in weekly review
- SYSTEM AND METHOD FOR VIDEO/AUDIO COMPREHENSION AND AUTOMATED CLIPPING
Set aside: nlp/localization
- Systems and methods for customizing channel programming
Set aside: business-ops
- Low Bit Rate Video Compression With Generative Video Models
Set aside: dropped in weekly review
- UNIFIED VIDEO QUALITY SCORE ASSESSMENT USING FUZZY LOGIC
Set aside: streaming, cdn/infra
- METHOD OF PREDICTING WATER ENVIRONMENTAL CONDITIONS
Set aside: streaming, nlp/localization, business-ops
- Tunable Hybrid Neural Video Representations
Set aside: codec
- Model-Based Visual Representation of Apparatus Operational Status
Set aside: cpc prior 0.06
- Generation of weights for causal inferences
Set aside: streaming
- Generation of cohort-agnostic features for models
Set aside: dropped in weekly review
- Image style transfer
Set aside: dropped in weekly review
- MEASUREMENT OF PROSODY SIMILARITY USING A PROSODY ENCODER
Set aside: dropped in weekly review
- LOGICAL WIDTH-BASED PROMPT CONDITIONING FOR GENERATIVE AI AND LLM
Set aside: streaming, recommendation, search/metadata
- Data Clustering for Machine Learning Model-Based Anomaly Prediction
Set aside: cpc prior 0.14
- Personalized Notification of Attractions Including User Preferences or Aversions
Set aside: dropped in weekly review
- MULTIMODAL MACHINE LEARNING MODEL FOR CONTENT EVALUATION
Set aside: streaming, recommendation
- Tagging performance evaluation and improvement
Set aside: search/metadata
- Dynamic mechanism for migrating traffic spikes in a streaming media network
Set aside: streaming, cdn/infra, security/drm
- Techniques for resampling images
Set aside: dropped in weekly review
- SYSTEM AND METHOD TO PROVIDE PERSONALIZED AUDIO STREAMING AND RENDERING
Set aside: streaming, codec, cdn/infra
- Visualization-Based Anomaly Detection for Predictive Maintenance
Set aside: dropped in weekly review
- METHOD AND APPARATUS FOR ENTITLEMENT
Set aside: security/drm
- QUERY DRAFTING VIA MACHINE LEARNING MODELS
Set aside: nlp/localization, business-ops
- PROFILE FILTERING BASED ON QUALITY OF SERVICE TARGET IN CONTENT DELIVERY
Set aside: streaming, codec, cdn/infra
- 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
Set aside: business-ops
- Performance optimization of pre-processor using video proxy codec
Set aside: codec
- ANOMALY DETECTION IN TIME SERIES DATA
Set aside: cdn/infra, business-ops
- Video Encoding Systems and Methods
Set aside: streaming, codec
- COVERAGE CONTROLLER FOR PACED DISTRIBUTION
Set aside: ad-tech
- SESSION TYPE CLASSIFICATION FOR MODELING
Set aside: cpc prior 0.09
- LARGE LANGUAGE MODEL TRANSLATOR
Set aside: dropped in weekly review
- ARTIFICIALLY INTELLIGENT AD-BREAK PREDICTION
Set aside: ad-tech, billing/fraud, cdn/infra
- Content matching tool for video streaming
Set aside: streaming, search/metadata
- Domain specific intelligence using natural language
Set aside: nlp/localization, business-ops
- COMBINING STATE SPACE MODELS AND CONVOLUTIONAL NEURAL NETWORKS FOR GENERIC VIDEO QUALITY ASSESSMENT
Set aside: streaming, codec, ad-tech
- Techniques for denoising videos
Set aside: dropped in weekly review
- System and method to remotely control and test media player devices
Set aside: streaming, ad-tech
- GENERATIVE METAMODEL FOR ACCELERATED SIMULATION MODELING
Set aside: dropped in weekly review
- SYSTEMS AND METHODS TO GENERATE PERSONALIZED LESSON CONTENT BASED ON USER CONTENT PREFERENCES
Set aside: dropped in weekly review
- METHOD AND SYSTEM FOR SERVER OVERLOAD PROTECTION
Set aside: cdn/infra
- AUTOMATED GENERATION OF COMMENTATOR-SPECIFIC SCRIPTS
Set aside: dropped in weekly review
- ASYNCHRONOUS REAL-TIME DELIVERY OF LOCALIZED STRINGS
Set aside: cdn/infra, nlp/localization
- Automated application security onboarding
Set aside: billing/fraud, security/drm, business-ops
- DYNAMIC MULTI-OBJECTIVE ASSIGNMENT OPTIMIZATION
Set aside: dropped in weekly review
- Intelligent content delivery network (CDN) entity routing
Set aside: cdn/infra
- Activity-based user interface personalization
Set aside: dropped in weekly review
- Artifact processing in video using texture information
Set aside: dropped in weekly review
- HOODED HANDLE ASSEMBLY OF A WATER VESSEL OR SNOW VESSEL
Set aside: cpc prior 0.41
- Priority-based production load balancing
Set aside: ad-tech, cdn/infra, business-ops
- Domain and user intent specific disambiguation of transcribed speech
Set aside: streaming, nlp/localization
- Tunable hybrid neural video representations
Set aside: codec
- Model-based visual representation of apparatus operational status
Set aside: cpc prior 0.06
- System and method for video/audio comprehension and automated clipping
Set aside: nlp/localization
- Multimodal machine learning model for content evaluation
Set aside: streaming, recommendation