AI Video Generation Instruction Storage

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for maintaining and regenerating AI-generated videos are inefficient, as they often require storing entire video files, which occupies significant storage space and does not allow for on-demand video generation.

Innovation Solution

A method that involves receiving user input to describe the requested video, processing this input using a generative AI model to generate detailed instructions for video creation, and then generating the video based on these instructions. Instead of storing the video file, the system saves the instructions, allowing for on-demand video regeneration.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If entire video files are stored for maintenance and regeneration, then video can be regenerated on-demand, but storage space consumption increases significantly

Engineering Contradiction:

Improvevideo regeneration capabilityVSAvoidstorage space

Core Design Contradiction:

Adaptability or versatilityVSVolume of stationary object

Solution Approach 1:

The patent extracts only the essential generation instructions and parameters from the complete video file, storing merely the metadata and control information rather than the full video content. This allows the system to regenerate videos on-demand while consuming minimal storage space, as only the instructional data is retained.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

Instead of storing the actual video files, the system creates and stores textual representations or descriptions of the videos in the form of generation instructions. These textual copies contain all necessary information to reconstruct the videos, eliminating the need for extensive storage of full video files.

Inventive Principle:

Principle #26Copying

2Productivity

If video files are stored for future generation, then on-demand video creation is enabled, but storage efficiency decreases

Engineering Contradiction:

Improveon-demand video generationVSAvoidstorage consumption

Core Design Contradiction:

ProductivityVSQuantity of substance

Solution Approach 1:

The patent segments the video generation process into distinct components: textual instructions, parameters, and metadata. Only these segmented instructional elements are stored, not the complete video. This segmentation allows efficient storage of minimal data while maintaining the ability to regenerate full videos on-demand.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system changes the state of video data from full binary video files to textual parameter representations. By transforming the storage format from complete video content to generation parameters and instructions, the system achieves high storage efficiency while preserving on-demand generation capability.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS20250139861A1System and method to maintaining generated ai video

Publication Date: 2025.05.01 IDOMOO LTD

AI summary

The present invention disclose: A method for maintaining generated Ai video, said method comprising the steps of:Receiving text message of user describing the requested video;Applying generative AI model for generating detailed instruction to generate the videoWherein the video is comprised of scenes, each scene has motion layout format including definitions of type of objects appearing, layout of objects, order of displaying objectsGenerating video based on detailed instruction to generate the videoOne user approve video saving detailed instruction to generate the video instead of the video enabling to generate the video ant any time