Effect Video Generation with Dynamic Point Cloud Planes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video application programs rigidly load effects to a single constructed plane based on 3D information, failing to meet the diversified demands of users and limiting the flexibility and personalization of effect addition in video processing.
Innovation Solution
Dynamically construct multiple planes in a three-dimensional space using point cloud data, determining distance information and data volume to apply a random sample consensus algorithm for plane estimation, allowing flexible and personalized effect addition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If effects are loaded to a single constructed plane based on 3D information, then the system structure is simple, but the flexibility and personalization of effect addition are limited
Solution Approach 1:
The patent segments the single plane into multiple planes in three-dimensional space. Each plane can independently承载 effects, allowing effects to be added to different spatial locations. This segmentation enables flexible and personalized effect addition while maintaining a relatively simple system structure through modular plane management.
Solution Approach 2:
The patent transitions from a two-dimensional single plane to a three-dimensional multi-plane structure. By introducing the third dimension (depth/spatial positioning), the system enables effects to be loaded onto multiple planes at different spatial positions, significantly improving flexibility and personalization without proportionally increasing system complexity.
2Adaptability or versatility
If a single plane is constructed from 3D information, then the processing is simple, but it cannot meet the diversified demands of users
Solution Approach 1:
The patent divides the single plane into multiple spatial planes, each capable of independently receiving effects. This allows users to add effects to different planes according to diverse demands, enhancing adaptability while keeping the operation process straightforward through automated plane selection based on point cloud data.
Solution Approach 2:
The patent implements dynamic plane construction where planes are created and managed based on real-time point cloud data analysis. The system dynamically determines which plane to load effects onto by calculating distance information and evaluating plane estimation conditions, making the effect adding operation adaptable to diverse user demands without complicating the user interface.
3Adaptability or versatility
If multiple planes are dynamically constructed in three-dimensional space, then the flexibility of effect addition is improved, but the computational complexity increases
Solution Approach 1:
The patent performs preliminary plane estimation by calculating distance information between point cloud data and historical planes before loading effects. This preliminary analysis allows the system to pre-determine the most suitable plane for effect loading, reducing real-time computational complexity while maintaining high flexibility in effect addition.
Solution Approach 2:
The patent implements a self-service mechanism where the system automatically evaluates plane estimation conditions and selects appropriate planes without requiring complex manual intervention. The random sample consensus algorithm and distance information calculation enable the system to autonomously manage multiple planes, improving flexibility while keeping the computational process streamlined through automated decision-making.
Data Source
AI summary
The present disclosure provides an effect video generation method and apparatus, an electronic device, and a storage medium. The effect video generation method includes: when it is detected that a current video frame to be processed includes point cloud data to be processed, determining distance information between the point cloud data to be processed and at least one historically estimated plane; and in a case that the distance information and a data volume of the point cloud data to be processed satisfy a plane estimation condition, performing plane estimation on the point cloud data to be processed on the basis of a random sample consensus algorithm to generate a target estimated plane, so as to display, based on each estimated plane, a target effect corresponding to the current video frame to be processed and obtain a target effect video frame.