Video Generation Using Cyclic Background Flipping for Dynamic Effects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video generation methods limit special effects and gameplay by only displaying the real scene, failing to meet users' demands for novelty and fun in video content.
Innovation Solution
A video generation method that involves acquiring background images, segmenting a target object from an image, and cyclically flipping these background images to create a dynamic display video by fusing the foreground and background images, allowing seamless blending of the target object with the background.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If only the real scene is displayed in videos, then the video generation process is simple, but the special effects and gameplay are limited
Solution Approach 1:
The video generation process is segmented into distinct modules: background image acquisition module, foreground object segmentation module, cyclic flipping module, and video fusion module. This segmentation allows complex video generation with special effects to be broken down into manageable steps, resolving the contradiction between versatility and complexity
Solution Approach 2:
The patent introduces dynamic cyclic flipping of background images and foreground objects at different frequencies and directions. This dynamic processing transforms static video capture into a versatile system that can generate various special effects and gameplay modes, enhancing adaptability while managing complexity through systematic design
2Adaptability or versatility
If background images are cyclically flipped to create dynamic scenes, then video fun and engagement increase, but processing complexity increases
Solution Approach 1:
The patent implements periodic cyclic flipping of background and foreground images at predetermined frequencies. This periodic action creates dynamic scene changes that enhance video engagement while optimizing processing energy consumption by avoiding continuous high-intensity operations, thus resolving the contradiction between scene dynamics and energy usage
Data Source
AI summary
The present disclosure provides a video generation method, an apparatus, a device and a medium, and the method includes: acquiring background images; acquiring an image to be segmented comprising a target object; performing image segmentation for the target object on the image to be segmented to obtain a contour image corresponding to the target object; using the contour image as a foreground image of a target display video; and cyclically flipping the background images, and generating the target display video by fusing the foreground image and the background images, in which the background images are being cyclically flipped in the target display video.