Video Node Location via Subtitle Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Users face inefficiencies in locating specific content within videos due to the lack of accurate and time-saving methods for navigating through rich content videos, such as knowledge or popular science videos, which require repeated viewing, as existing methods like fast-forwarding or previewing frames are not precise enough.

Innovation Solution

A method and apparatus that select a target video based on value measurement, display subtitle text segments with start time information, and allow users to determine a start playing node by triggering a specific subtitle text segment, thereby saving computing resources and improving interaction efficiency between the server and client.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If users use fast-forwarding or previewing frames to locate specific content in videos, then users can navigate through videos, but the locating precision is insufficient and time-consuming

Engineering Contradiction:
Improvelocating precisionVSAvoidtime for locating content
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the video content by extracting subtitle text and dividing it into multiple text segments, each associated with specific time information. This segmentation allows users to locate specific content by searching through discrete text segments rather than continuously scanning video frames, significantly improving locating precision and reducing time consumption.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If the system processes and analyzes all video content to enable precise locating, then locating accuracy improves, but computing resources are consumed

Engineering Contradiction:
Improvecontent locating accuracyVSAvoidcomputing resource consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent extracts only the essential subtitle text from the video content, separating this key information from the full video data. By processing and indexing only the extracted text segments rather than the entire video, the system achieves accurate content locating while significantly reducing computing resource consumption.

Inventive Principle:
Principle #2Taking out (Extraction)

3Productivity

If the system displays full subtitle text to help users locate content, then users can find specific information faster, but interaction complexity increases

Engineering Contradiction:
Improvecontent finding speedVSAvoidinteraction complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent applies local quality by displaying different portions of subtitle text with different visual characteristics. Specific text segments corresponding to searchable content are highlighted or distinguished from other parts, allowing users to quickly identify relevant content without being overwhelmed by the full subtitle text, thus improving content finding speed while maintaining simple interaction.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3855753B1Method and apparatus for locating video playing node, device and storage medium
Publication Date: 2023.03.15 BEIJING BAIDU NETCOM SCI & TECH CO LTD
  • EP3855753B1 patent drawingFigure 1~2A
  • EP3855753B1 patent drawingFigure 2B
  • EP3855753B1 patent drawingFigure 3~4

AI summary

The disclosure provides a method for locating a video playing node, and relates to fields of big data and video processing. The method includes: selecting a target video out from a plurality of videos; and sending the target video, a plurality of subtitle text segments of the target video and start time information of each of the plurality of subtitle text segments to a client, to cause the client to display the plurality of subtitle text segments, and determine, in response to a trigger operation on an any subtitle text segment of the plurality of subtitle text segments, a start playing node of the target video based on the start time information of the any subtitle text segment. The disclosure further provides an apparatus for locating a video playing node, an electronic device and a storage medium.