Iterative Bone Weight Matrix Update for Vertex Animation Conversion

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The existing methods for converting vertex animation videos to bone animation videos face challenges in accurately determining the bone weight matrix, leading to low fitting accuracy due to differences between objects in various videos.

Innovation Solution

A video processing method that iteratively updates the bone weight and rotation/translation matrices based on the differences between predicted and actual vertex positions, improving the accuracy of the bone weight matrix by alternating between updating these matrices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If template matching method is used to determine the bone weight matrix, then the conversion from vertex animation video to bone animation video can be achieved, but the fitting accuracy is low due to differences between objects in different videos and templates

Engineering Contradiction:
Improvefitting accuracy of bone weight matrixVSAvoidadaptability to different objects
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent changes the parameters of the bone weight matrix through iterative optimization. Instead of using a fixed template matching approach, the system adjusts the weight values dynamically by minimizing the difference between predicted and actual vertex positions, thereby adapting to different objects while improving fitting accuracy

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent implements a feedback mechanism where the predicted vertex positions are compared with actual positions, and the bone weight matrix is updated based on the error difference. This closed-loop optimization continues iteratively until convergence, enabling the system to adapt to various objects while maintaining high fitting accuracy

Inventive Principle:
Principle #23Feedback

2Manufacturing precision

If more vertices are used to record relevant information of each video frame, then the video effect (clarity and fluency) is improved, but the memory overheads required for the vertex animation video increase

Engineering Contradiction:
Improvevideo effect qualityVSAvoidmemory overhead
Core Design Contradiction:
Manufacturing precisionVSQuantity of substance

Solution Approach 1:

The patent extracts the essential animation information by converting from a detailed vertex-based representation to a more compact bone-based representation. By taking out only the critical structural elements (bones and their weights), the system maintains video quality while significantly reducing memory requirements

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies local quality by using different levels of detail in different parts of the model. The bone weight matrix provides precise control where needed (at bone-vertex relationships) while using a simplified hierarchical structure elsewhere, achieving high quality output with reduced overall data requirements

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS20240394947A1Video processing method and apparatus, device, storage medium, and product
Publication Date: 2024.11.28 TENCENT TECHNOLOGY (SHENZHEN) CO LTD
  • US20240394947A1 patent drawing
  • US20240394947A1 patent drawing
  • US20240394947A1 patent drawing

AI summary

This application disclose a video processing method performed by a computer device. The method includes: obtaining a bone rotation and translation matrix of an object in a first video frame of a target video, predicting positions of the M vertices in a second video frame of the target video subsequent to the first video frame by using a bone weight matrix of the object and the bone rotation and translation matrix, to obtain predicted positions of the M vertices; alternatively updating the bone weight matrix and the bone rotation and translation matrix based on a difference between real positions of the M vertices in the second video frame and the predicted positions of the M vertices, to obtain a target bone weight matrix; and restoring the target video based on the target bone weight matrix and the bone rotation and translation matrix of the object in the target video.