Face Recognition Video Clipping for Dynamic Game Streaming

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In moving image distribution services, effectively showcasing a user's state, including their face, is challenging, as existing technologies lack efficient methods for personalized and dynamic camera video distribution during game play.

Innovation Solution

An information processing device with an image obtaining section, face recognizing section, display user determining section, face image clipping section, and display image generating section, which detects and clips face images from camera video to generate and distribute personalized display images based on user instructions, allowing real-time face recognition and dynamic camera video superimposition on game screens.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the camera video of multiple users is distributed simultaneously, then the viewer engagement is improved, but the device complexity and processing load increase

Engineering Contradiction:
Improveviewer engagementVSAvoidprocessing load
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent extracts only the face region from the full camera video feed using face recognition technology. Instead of transmitting and processing entire video streams of multiple users, the system identifies facial boundaries and extracts only the relevant face portions, significantly reducing data volume and processing requirements while maintaining viewer engagement through personalized face display.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system creates simplified copies of the original video data by generating face-only representations. Rather than handling complete video streams, the patent generates cropped face images or face-region video segments that can be easily transmitted and processed, reducing the computational burden while preserving the essential visual information for viewer engagement.

Inventive Principle:
Principle #26Copying

2Measurement precision

If the face recognition and dynamic selection processing is performed in real-time, then the personalization quality is improved, but the processing time and computational resources increase

Engineering Contradiction:
Improvepersonalization qualityVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs face recognition and identifies target users before the actual video distribution process. By pre-processing the video streams to detect and mark face regions in advance, the system prepares the data for rapid transmission and display. This preliminary face detection and region marking reduces the processing time during live distribution while maintaining high personalization quality.

Inventive Principle:
Principle #10Preliminary action

3Loss of information

If the entire camera video is transmitted to viewers, then the information completeness is improved, but the data transmission volume and network bandwidth consumption increase

Engineering Contradiction:
Improveinformation completenessVSAvoiddata transmission volume
Core Design Contradiction:
Loss of informationVSQuantity of substance

Solution Approach 1:

The patent extracts only the face region from the complete camera video feed for transmission to viewers. By using face recognition to identify and extract facial boundaries, the system transmits only the relevant face portions rather than entire video streams, significantly reducing data transmission volume while preserving the essential visual information that viewers need for engagement.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentUS11109108B2Information processing device
Publication Date: 2021.08.31 SONY INTERACTIVE ENTERTAINMENT LLC
  • US11109108B2 patent drawing
  • US11109108B2 patent drawing
  • US11109108B2 patent drawing

AI summary

An imaged image obtaining section obtains an imaged image from a camera. A face recognizing section detects the face images of a plurality of users in the imaged image. A display user determining section has a function of determining a user to be included in a display image. When an instruction receiving section receives a changing instruction, the display user determining section changes a user included in the display image. A face image clipping section clips a region including the face image of the determined user from the imaged image. A display image generating section generates the display image including the clipped region.