Spectator Switchboard Viewport Selection via Machine Learning

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Remote spectators of live gaming events often find it inefficient and uninteresting to navigate through numerous content options, leading to disengagement due to the complexity and lack of relevant content alignment with their preferences.

Innovation Solution

A system that uses machine learning models to dynamically select and present viewports based on spectator profiles and live event data, providing a customized switchboard interface with predicted interesting actions, allowing efficient navigation and automatic recording of non-selected content for later viewing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If spectators manually navigate through multiple content options in live gaming events, then they can access various viewpoints, but the complexity of navigation and time required to find interesting content increases

Engineering Contradiction:
Improveviewport selectionVSAvoidnavigation complexity
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

The system automatically selects and presents relevant viewports to spectators based on their profile data and real-time game event analysis, eliminating the need for manual navigation. The machine learning model performs self-service by autonomously curating content that matches spectator preferences, thereby reducing navigation complexity while maintaining adaptability.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent introduces an intermediary system comprising a machine learning model and switchboard interface that mediates between the available game content and the spectator. This intermediary automatically processes multiple content options and presents only the most relevant viewports, simplifying the spectator's task while preserving access to diverse viewpoints.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of information

If spectators manually search through content to find interesting scenes, then they can locate desired viewports, but the time required to find relevant content increases

Engineering Contradiction:
Improvecontent discoveryVSAvoidsearch time
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

The system performs preliminary action by pre-processing game event data and spectator profile information before the spectator needs to view content. The machine learning model analyzes upcoming game events and proactively identifies interesting scenes, preparing customized viewport recommendations in advance. This eliminates the need for spectators to manually search through content, significantly reducing search time while ensuring relevant content is delivered.

Inventive Principle:
Principle #10Preliminary action

3Adaptability or versatility

If a customized switchboard interface is generated using machine learning models, then viewport relevance to spectator preferences is improved, but system complexity increases

Engineering Contradiction:
Improvecontent personalizationVSAvoidsystem architecture
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the complex system into distinct functional modules: a machine learning model for analyzing spectator profiles and game events, a switchboard interface for presenting viewports, and separate processing components for different types of data. This segmentation manages system complexity by organizing functions into independent, manageable units while maintaining high adaptability and personalization capabilities.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS11612818B2Spectator switch board customized to user viewport selection actions
Publication Date: 2023.03.28 SONY INTERACTIVE ENTERTAINMENT LLC
  • US11612818B2 patent drawing
  • US11612818B2 patent drawing
  • US11612818B2 patent drawing

AI summary

Methods and systems are provided for selecting viewports into a game. One example method includes identifying a plurality of virtual cameras for providing viewports into the game. The method includes accessing a playbook of a spectator user, and the playbook identifies performance of the spectator user in the game. The method includes accessing event data for the game as the game is played live or recorded. The method includes generating a switchboard interface for the spectator user including a plurality of viewports providing optional views into the game. The plurality of viewports is dynamically selected based on processing the event data and the playbook of the spectator user through a machine learning model. The machine learning model is configured to identify features from the event data and the playbook to identify attributes of the spectator user. The attributes of the spectator user are used to select the plurality of viewports into the game.