Captioned Video Scene Tracking for User Engagement Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video analysis methods struggle to provide detailed insights into which scenes of a video are viewed by users, leading to vague data analysis and difficulty in evaluating user engagement effectively.

Innovation Solution

A video viewing providing device that sets scenes with captions and checks user viewing status, generating data on scene viewing times, and provides feedback to administrators for evaluation and marketing purposes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If video viewing analysis is performed using conventional methods, then the system is simple to operate, but the measurement precision of user engagement is insufficient

Engineering Contradiction:
Improvescene viewing analysis precisionVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The video is divided into multiple scenes with captions, and the system tracks viewing status for each scene separately. This segmentation enables precise measurement of which specific scenes users watched, rather than treating the video as a single unit, thereby improving measurement precision without requiring complex hardware modifications

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

A scene data file is introduced as an intermediary component that stores scene information (captions, timecodes) separately from the video content. This intermediary structure allows the system to track scene-level viewing data without complicating the core video delivery mechanism, resolving the contradiction between precision and complexity

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of information

If detailed scene viewing data is collected for all users, then the information value for marketing increases, but the quantity of data to be processed increases

Engineering Contradiction:
Improveuser engagement information valueVSAvoiddata volume
Core Design Contradiction:
Loss of informationVSQuantity of substance

Solution Approach 1:

The system extracts and stores only essential scene viewing information (which scenes were watched, viewing duration) in a structured scene data file, rather than collecting and processing all possible video metadata. This extraction approach maintains high information value for marketing while minimizing data volume by focusing only on relevant engagement metrics

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentUS12418698B2Video viewing providing device, video viewing providing method, and video viewing providing program
Publication Date: 2025.09.16 INFORM SYST INC
  • US12418698B2 patent drawing
  • US12418698B2 patent drawing
  • US12418698B2 patent drawing

AI summary

[Problem]To provide a video viewing providing device, a video viewing providing method, and a video viewing providing program that, during video delivery, allow for visualization of which scene has been viewed by each user, and make it possible to easily evaluate a video or the user who has viewed the video.[Solution]A video viewing providing device 100 comprises: a video delivery unit 10 for delivering a video to a plurality of users; a scene setting unit 20 for setting scenes with a caption at a plurality of locations of the video; a viewing evaluating unit 30 for checking whether the user has viewed a location where a scene with a caption is set; and a feedback unit 40 for feeding back an evaluating result of the viewing evaluating unit to the user and a delivery source of the video. The scene setting unit 20 generates data of a plurality of scenes with a caption including captions and a start time and an end time in the time line of the video.