AV Stream Feature Detection for Automatic Playlist Generation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Users face difficulties in creating a playlist of characteristic moving image content from recorded AV streams without manually re-checking the content, as existing methods require significant user input to identify and edit playback segments.

Innovation Solution

An information processing apparatus that detects feature information from AV streams, sets playback segments based on detection time points, and generates playback-segment data to automatically create a playlist, allowing users to quickly extract and edit interesting segments without reviewing all content.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If manual review of captured moving image content is performed to specify characteristic portions, then playback lists can be created with user input, but the process requires significant time and user effort

Engineering Contradiction:
Improveease of playlist creationVSAvoidtime for content review
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The system automatically detects feature information in captured moving images and generates playback lists without requiring user review. The playback-segment setting section automatically sets playback segments based on detected feature information, enabling the system to serve itself rather than requiring manual user input for playlist creation

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system performs preliminary detection of feature information during or immediately after capture, preparing playback-segment data in advance. This preliminary action allows the playback list to be generated automatically without requiring subsequent manual review by the user

Inventive Principle:
Principle #10Preliminary action

2Extent of automation

If automatic detection of feature information is implemented, then playlist creation becomes automated, but the system complexity increases

Engineering Contradiction:
Improveautomation of playlist creationVSAvoidsystem complexity
Core Design Contradiction:
Extent of automationVSDevice complexity

Solution Approach 1:

The system divides the complex task of automatic playlist creation into distinct functional sections: a detecting section for feature information detection, a playback-segment setting section for segment configuration, and a data generating section for playlist generation. This segmentation allows each component to handle a specific subtask, reducing overall system complexity while achieving high automation

Inventive Principle:
Principle #1Segmentation

3Productivity

If playback segments are set based on detection time points, then characteristic portions are automatically extracted, but precision in segment boundaries may be compromised

Engineering Contradiction:
Improvespeed of content processingVSAvoidprecision of segment boundaries
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The system applies different processing qualities to different parts of the content. Feature information detection focuses on identifying characteristic portions with high precision, while playback segment boundaries are set locally around these detected features. This allows the system to maintain high productivity by processing content based on detected features rather than uniform sampling, while preserving sufficient precision at the segment boundaries where it matters most

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS8886012B2Information processing apparatus, information processing method, and program
Publication Date: 2014.11.11 SONY GROUP CORP
  • US8886012B2 patent drawing
  • US8886012B2 patent drawing
  • US8886012B2 patent drawing

AI summary

An information processing apparatus includes: a detecting section that detects feature information of at least one AV stream; a recording section that records the AV streams, clip information defining attributes including the detected feature information, and playback-segment data specifying at least one playback segment to a storage medium in accordance with a predetermined data-recording format, the playback-segment data being obtained by setting a playback start point and a playback end point with respect to the AV streams; a playback-segment setting section that sets the playback segments with respect to the AV streams on the basis of detection time point of the feature information; and a data generating section that generates the playback-segment data specifying the playback segments set by the playback-segment setting section.