Stats LLC

Disney Enterprises, Inc.

Patrick Joseph Lucey

Sujoy Ganguly

Xinyu Wei

Long Sha

Aditya Cherukumudi

Jennifer Hobbs

Iain Matthews

Paul David Power

Christian Marko

Hector Ruiz

Joe Dominic Gallagher

Panna Felsen

George Peter Carr

Matthew Scott

Michael Stöckl

Alina Bialkowski

Matthew Holbrook

Daniel Richard Dinsdale

Yaser Sheikh

Sateesh Pedagadi

Eric Foote

Alex Ottenwess

Daniel Candela

Johannes Kuehnel

Nathan Frank

Raphael Reiners

Brian Orefice

Thomas Seidl

Ken Rhodes

G Peter Carr

Rajitha Navarathna

William Thomas Gurpinar-Morgan

Bradford Griffiths

Kevin Allinson

Anthony Borsumato

Robert Seidl

G Peter K Carr

Yisong Yue

Filip Glojnaric

Matjaz Ales

Christopher Evi-Parker

Sateesh Padagadi

Adedapo Alabi

Akila Pemasiri

Nils Sebastiaan Mackaij

Alexander Nicholas Ottenwess

Ben Mackriell

Ysabel Gonzalez-Rico

Arun Murali

Valerio Colamatteo

Peter Carr

Examples disclosed herein may estimate locations of players not visible in a sporting broadcast video. A prediction model may be generated based on a training data set of in-venue tracking data that includes locations of all players at all times and the corresponding broadcast tracking data that may not necessarily contain the locations of all players at all times. The prediction model may be based on an algorithmic logic (e.g., a spline regression) or machine learning model (e.g., k-nearest neighbor, deep neural network). The generated predicted model may be used to estimate the unknown locations of players in broadcast tracking based on the known locations.

Estimating missing player locations in broadcast video feeds

A system and method for predicting multi-agent locations is disclosed herein. A computing system retrieves tracking data from a data store. The computing system generates a predictive model using a conditional variational autoencoder. The conditional variational autoencoder learns one or more paths a subset of agents of the plurality of agents are likely to take. The computing system receives tracking data from a tracking system positioned remotely in a venue hosting a candidate sporting event. The computing system identifies one or more candidate agents for which to predict locations. The computing system infers, via the predictive model, one or more locations of the one or more candidate agents. The computing system generates a graphical representation of the one or more locations of the one or more candidate agents.

System and method for predicting fine-grained adversarial multi-agent motion

A computing system retrieves historical event data for a plurality of games in a league. The historical event data includes (x,y) coordinates of players within each game and game context data. The computing system learns one or more attributes of each team in each game and each player on each team in each game. The computing system receives a request to simulate a play in a historical game. The request includes substituting a player that was in the play with a target player that was not in the play. The computing system simulates the play with the target player in place of the player based on the one or more attributes learned by the computing system. The computing system generates a graphical representation of the simulation.

System and method for individual player and team simulation

A computing system receives a broadcast video stream of a game. A codec module of the computing system extracts image level features from the broadcast video stream. The codec module includes an object detection portion configured to detect players in the broadcast video stream and a subnet portion attached to the object detection portion. The subnet portion is configured to identify foreground information of the detected players. The codec module provides the image level features to a plurality of task specific modules for analysis. The plurality of task specific modules generates a plurality of outputs based on the image level features.

Sports neural network codec

A computing system receives a plurality of game files corresponding to a plurality of games across a plurality of seasons. The computing system generates a prediction model configured to generate a possession value for an event. The computing system receives a target event, in real-time or near real-time, from a tracking system monitoring a target game. The computing system generates target features for the target event based on target event data associated with the target event. The computing system generates, via the prediction model, a target possession value for the target event based on the target event data and the target features. The target possession value represents a likelihood that a team with possession will score within a following x-seconds after the target event.

Live possession value model

A computing system receives pre-match data for an upcoming match between a first player and a second player. The computing system generates, using one or more prediction models, one or more pre-match predictions based on the pre-match data. The computing system receives in-match data for the match currently in progress. The computing system generates, using the one or more prediction models, one or more live match predictions based on the in-match data.

Live prediction of player performances in tennis

A system and method of predicting a team's formation on a playing surface are disclosed herein. A computing system retrieves one or more sets of event data for a plurality of events. Each set of event data corresponds to a segment of the event. A deep neural network, such as a mixture density network, learns to predict an optimal permutation of players in each segment of the event based on the one or more sets of event data. The deep neural network learns a distribution of players for each segment based on the corresponding event data and optimal permutation of players. The computing system generates a fully trained prediction model based on the learning. The computing system receives target event data corresponding to a target event. The computing system generates, via the trained prediction model, an expected position of each player based on the target event data.

System and method for predicting formation in sports

A method of generating a player prediction is disclosed herein. A computing system retrieves data from a data store. The computing system generates a predictive model using an artificial neural network. The artificial neural network generates one or more personalized embeddings that include player-specific information based on historical performance. The computing system selects, from the data, one or more features related to each shot attempt captured in the data. The artificial neural network learns an outcome of each shot attempt based at least on the one or more personalized embeddings and the one or more features related to each shot attempt.

Personalizing prediction of performance using data and body-pose for analysis of sporting performance

A method for generating a probability for a first action of a sporting event by implementing a feature set, the method including: obtaining an initial set of data relating to the first action of a sporting event, the initial set of data including at least a position of a first player on a surface and a position of a target area on the surface; generating, by a machine learning model, an initial projected scoring probability based on the initial set of data; generating a feature set relating to the sporting event; and modifying, by the machine learning model, the initial projected scoring probability to an updated scoring probability using the feature set.

System and methods for implementing a feature set of high-dimensional spatial data in sports predictions

A method of generating a multi-modal prediction is disclosed herein. A computing system retrieves event data from a data store. The event data includes information for a plurality of events across a plurality of seasons. Computing system generates a predictive model using a mixture density network, by generating an input vector from the event data learning, by the mixture density network, a plurality of values associated with a next play following each play in the event data. The mixture density network is trained to output the plurality of values near simultaneously. Computing system receives a set of event data directed to an event in a match. The set of event data includes information directed to at least playing surface position and current score. Computing system generates, via the predictive model, a plurality of values associated with a next event following the event based on the set of event data.

System and method for multi-task learning

Examples disclosed herein may generate a refined and denoised body pose data from a video feed of a sporting event. Tracking data containing player locations may be used to determine correspondence between a location and a body pose. For example, body pose with middle of key footpoints with shortest distance from the location may be selected as a likely body pose for the location. The body pose data may be refined to estimate the length of missing limbs or limbs with unusual length ratios. The body pose data may further be filtered to filter out unwanted body poses such as body poses of spectators or noisy body poses. The refined and filtered body pose data may be used for other downstream processing such as projecting the body poses to a three dimensional play surface.

Body pose tracking of players from sports broadcast video feed

A system and method of calibrating moving cameras capturing a sporting event is disclosed herein. A computing system retrieves a broadcast video feed for a sporting event. The broadcast video feed includes a plurality of video frames. The computing system labels, via a neural network, components of a playing surface captured in each video frame. The computing system matches a subset of labeled video frames to a set of templates with various camera perspectives. The computing system fits a playing surface model to the set of labeled video frames that were matched to the set of templates. The computing system identifies camera motion in each video frame using an optical flow model. The computing system generates a homography matrix for each video frame based on the fitted playing surface model and camera motion. The computing system calibrates each camera based on the homography matrix generated for each video frame.

System and method for calibrating moving cameras capturing broadcast video

A computing system receives data for a game. The data includes at least one of tracking data or event data. Based on the data for the game, the computing system determines that an event has occurred within the game. Based on the determining, the computing system generates a graphic responsive to the event. The graphic includes insights related to the event. The computing system recommends an image relevant to the event based on metatags associated with the event. The computing system generates a visual element by merging the image and the graphic.

Recommendation engine for combining images and graphics of sports content based on artificial intelligence generated game metrics

A system and method of re-identifying players in a broadcast video feed are provided herein. A computing system retrieves a broadcast video feed for a sporting event. The broadcast video feed includes a plurality of video frames. The computing system generates a plurality of tracks based on the plurality of video frames. Each track includes a plurality of image patches associated with at least one player. Each image patch of the plurality of image patches is a subset of the corresponding frame of the plurality of video frames. For each track, the computing system generates a gallery of image patches. A jersey number of each player is visible in each image patch of the gallery. The computing system matches, via a convolutional autoencoder, tracks across galleries. The computing system measures, via a neural network, a similarity score for each matched track and associates two tracks based on the measured similarity.

System and method for player reidentification in broadcast video

A computing system identifies player tracking data and event data corresponding to a match. The match includes a first team and a second team. The player tracking data includes coordinate positions of each player during the event. The event data defines events that occur during the match. The computing system divides the player tracking data into a plurality of segments based on the event information. For each segment of the plurality of segments, the computing system learns a first formation associated with a respective team in possession. For each segment of the plurality of segments, the computing system learns a second formation associated with a respective team not in possession. The computing system maps each first formation to a first class of known formation clusters. The computing system maps each second formation to a second class of known formation clusters.

Interactive formation analysis in sports utilizing semi-supervised methods

Growing community of inventors

Chicago, IL, United States of America

Patrick Joseph Lucey