System and Method for Detection and Visulaization of Anomalous Media Events

Number: 20180095652
Published: 2018-04-05
Filed: 2016-09-30
Assignee: Disney Enterprises, Inc.
Inventors: Branch; Joel et al.
CPC: G08B25/008; G06F3/04847; G08B5/36; H04L51/214; H04L51/222; H04L51/52
Verdict: Set aside anomalous media-event detection, broadcast monitoring/business
Source: Google Patents · FreePatentsOnline

Abstract

Methods and systems for detection and visualization of anomalous media events are described, where AV media data of a sporting event (game) are analyzed to identify AV media segments indicative of unscripted anomalous events (AE's) that elicit a social media response (SMR), and selectable AE graphic visualization (or AE GUI) is provided showing the position of the AE on the court or field of play graphic, corresponding to the actual field of play, as a form of augmented reality. The graphic may also show the level of interest and the sentiment allocation (positive, negative, neutral) for each AE, a social media source allocation, details about the AE, social media posts, and other graphic display features related to the AE. Feedback may also be provided to adjust the logic that identifies AE. AE Alerts may also be sent to users autonomously or based on user-set parameters.

Background

BACKGROUND

Facilitating the querying and indexing of media content, such as video and audio, is a process that requires extensive time due to manual processes. Such media querying and indexing is currently done via manual inspection and review of the media content as one traverses it forward and backward in time. In addition, manual annotation of points in time (e.g., embedding text descriptions of scenes within a video) is sometimes performed to enable faster and more descriptive querying. This is adequate for applications where the annotations will not change over time (e.g., for a movie or recorded stage production) or where the annotations mainly pertain to descriptions or interpretations of the actual media.

However, querying and indexing of unscripted media, such as sports media, or other media that is naturally the subject of viewer opinion (or sentiment), has additional requirements. In particular, consumers also have a desire to view segments of sports media that are associated with or connected to, or that elicit, result in, cause, or generate, crowd (or general viewing public) reaction, response, opinion, or emotion, especially since sports content viewing is often a communal event. Additionally, such crowd-related data may change over time or may vary based on type of event, sport, action, location, or other factors or variables.

Accordingly, it would be desirable to have a method and system that provides indexing and querying of sports (or simil

Claims

1. A method for identifying at least one anomalous event (AE) within a sporting event and identifying and displaying the location of the AE and a social media response to the AE, comprising: receiving AV media data indicative of the sporting event; identifying the AE within the AV media data; determining the location of the AE on the actual field of play; receiving social media data; identifying at least one social media response to the AE within the social media data; and providing a graphic user interface (GUI) showing a virtual version of the field of play and at least one AE graphic on the virtual field of play, the AE graphic indicative of the location of the AE and at least one social media parameter associated with the social media response to the AE. 24. A method for identifying at least one anomalous event (AE) within a sporting event, comprising: receiving AV media data indicative of the sporting event; analyzing the AV media data for at least one factor related to at least one of: crowd noise, commentator voice intonation, commentator language/words, AE attributes, and influencer sources language/words; and identifying the occurrence of an AE based on an AE identification model, the model having model parameters based on the at least one factor. 30. A method for identifying at least one anomalous event (AE) within a sporting event and identifying and displaying the location of the AE and a social media response to the AE, comprising: receiving AV media data indicative of the sporting event; identifying the AE within the AV media data; receiving social media data; identifying at least one social media response to the AE within the social media data; and providing a graphic user interface (GUI) showing an AE timeline and at least one AE graphic on the AE timeline, the AE graphic indicative of at least one social media parameter associated with the social media response to the AE.