A video production method and device based on video script semantic recognition

Abstract

The present application provides a video production method and device based on semantic recognition of video scripts. The method includes obtaining a video production script given by the user. The video production script carries a plurality of video production sentences, and each video production sentence Including a material type identification tag and a material attribute tag, the material attribute tag includes a material description and / or a material display form; the material library for material selection is determined according to the material type identification tag, and in the material library, according to the The material attribute tag determines the material and / or determines the operation mode of the material, and generates a video from the material according to the order of the video production statements in the video production script. Through the above method or device, the present application can intelligently analyze massive video materials, combine semantic understanding of video scripts, and automatically generate short videos.

Description

Technical field [0001] The present application belongs to the field of video processing, and it is particularly related to a video production method and apparatus based on video script semitism. Background technique [0002] The traditional short video artificial production method mainly includes the following steps: 1, content planning, write video scripts; 2, shooting material or editing existing content as available materials; . Each step of this process requires a relatively large human resources and time inside. [0003] The existing video template is automatically produced, that is, in prior to editing a set of fixed video materials for a particular scenario, then the user entered the specified text content, subtitle and text display content and voice replacement. In this way, the video form produced is basically fixed, except for the text content part, the video picture content is relatively fixed. It is equivalent to the video script is a fixed mode and the material can o...

AI Technical Summary

Problems solved by technology

Although this method solves part of labor and time costs, the application scenarios have limitations, but for some custom video scripts, this method cannot be used to solve the consumption of manpower and time

For example, templates related to traffic accidents can only be limited to the production of videos related to traffic accidents. If you want to produce videos with other content, one way is for system developers to add related templates, which increases the work of developers, or by Video editors manually replace the material in the template, which increases the workload of video editors

Images