一种基于主体识别的横屏转竖屏故事板智能重构方法及系统

CN122420596APending Publication Date: 2026-07-17北京广播电视台 +1

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
北京广播电视台
Filing Date
2026-04-02
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing technologies suffer from low efficiency, difficulty in guaranteeing quality, insufficient data security, and disruption of professional editing processes during the conversion of landscape videos to portrait videos. In particular, in video processing involving trade secrets or personal privacy, it is impossible to balance high efficiency and automation with data security.

Method used

Employing locally deployed subject recognition technology, the system dynamically adjusts the cropping area and scaling ratio through a video/image saliency detection model, generating a vertical storyboard project file containing keyframe data. This allows for manual adjustments by the user, enabling intelligent reconstruction of adaptive composition and visual effects.

Benefits of technology

It improves conversion efficiency, enhances visual quality, ensures data security, provides professional editing flexibility and editability, and adapts to the production needs of various video content types.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122420596A_ABST
    Figure CN122420596A_ABST
Patent Text Reader

Abstract

本发明公开一种基于主体识别的横屏转竖屏故事板智能重构方法及系统,所述方法包括以下步骤:步骤1、编辑完成横屏故事板工程;步骤2、初始化竖屏故事板工程;步骤3、调用本地模型进行主体识别与画面重构;步骤4、生成并嵌入重构关键帧数据;步骤5、字幕文件自动适配与转换;步骤6、重构完成竖屏故事板工程。所述系统包括相连接的工程编辑模块、本地视觉模型模块、画面重构算法模块、关键帧数据模块、字幕适配及编辑模块。本发明实现了从横屏故事板工程到竖屏故事板工程的智能化重构,能够大幅提升内容转化效率,显著降低人力与时间成本;并且提升视觉质量,保障数据安全,同时具备较强的可编辑性,场景适应灵活。
Need to check novelty before this filing date? Find Prior Art