An automated video clip system based on matching of text script and picture content

By using a matching technology based on text scripts and visual content, the system accurately quantifies the narrative fluctuation rate and audio rhythm of the script, establishes a quantitative index of visual tension in the material, and constructs a state transition matrix to verify the consistency between the visual tension difference and the sign of the script's emotional derivative. This solves the problem of inaccurate matching in existing video editing systems and achieves highly efficient video editing results.

CN122120547APending Publication Date: 2026-05-29SHIJIAZHUANG YINGDAO CULTURE MEDIA CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHIJIAZHUANG YINGDAO CULTURE MEDIA CO LTD
Filing Date
2026-01-28
Publication Date
2026-05-29

AI Technical Summary

Technical Problem

Existing automated video editing systems cannot accurately match the emotions and narrative rhythm of video footage with text scripts, resulting in a disconnect between video content and text descriptions, and low editing efficiency.

Method used

By collecting text scripts to map emotional intensity curves and calculating first-order differences, combining audio amplitude envelope mutation detection to divide atomic time intervals, establishing a visual tension quantification index for the material, constructing a state transition matrix to verify the consistency between the visual tension difference and the sign of the script's emotional derivative, determining the transition mode and performing nonlinear resampling to adapt the time slots, and generating script matching for automated editing into a film.

Benefits of technology

It achieves deep emotional coupling between video and text, synchronizes audiovisual rhythm, and enhances the narrative appeal and fluency of the video.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122120547A_ABST
    Figure CN122120547A_ABST
Patent Text Reader

Abstract

The application relates to the technical field of video clip, in particular to an automatic video clip system based on text script and picture content matching, which comprises a script time domain analysis module, a material kinetic energy extraction module, a clip path planning module, a transition special effect injection module and a film rendering synthesis module.In the application, the script emotion curve is mapped and first-order difference is calculated, the audio mutation detection is combined to divide an atomic time interval, the narrative fluctuation is quantified to realize audio-visual emotion depth coupling, the optical flow vector and the brightness variance are calculated to establish kinetic energy mapping, the picture is converted into a measurable tension index, the state matrix is constructed to check the consistency of the tension difference value and the emotion derivative sign, the reverse path is removed to ensure that the picture kinetic energy and the narrative trend vector are in the same direction, the transition is matched according to the semantic distance and the motion angle, and nonlinear resampling is performed, so that the optical flow transition is smooth, the audio beat is accurately synchronized, the narrative infectivity of the film is improved, and the audio-visual fluency is improved.
Need to check novelty before this filing date? Find Prior Art