用于视频生成的方法、装置、设备、存储介质和程序产品
By using machine learning models to generate segmentation results with identification information and text descriptions for video clips, the problem of low efficiency and poor quality in existing video editing technologies is solved, and efficient and high-quality video generation is achieved.
CN119629439BActive Publication Date: 2026-07-17BEIJING YOUZHUJU NETWORK TECH CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING YOUZHUJU NETWORK TECH CO LTD
- Filing Date
- 2024-12-11
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Existing video editing methods are inefficient and produce low-quality content; there is a need for efficient video editing methods that can generate high-quality content.
Method used
Using a trained machine learning model, a target file is generated based on multiple video clips and prompts. The target file includes the identification information, text description, and segmentation results of the video clips, thereby generating the target video.
Benefits of technology
It improves the efficiency of video editing and enhances the quality of generated videos.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN119629439B_ABST
Abstract
本公开的实施例提供了用于视频生成的方法、装置、设备、存储介质和程序产品。该方法包括：获取针对目标对象的多个视频片段和目标对象的对象信息，多个视频片段分别具有对应的第一标识信息；利用经训练的机器学习模型，基于多个视频片段和提示词生成目标文件，提示词至少包括对象信息和多个视频片段各自的第一标识信息，并且目标文件至少包括与多个视频片段中的按顺序排列的至少一个视频片段对应的第一标识信息、对至少一个视频片段的文本描述以及对文本描述的分割结果；以及至少基于目标文件，生成针对目标对象的目标视频。
Need to check novelty before this filing date? Find Prior Art