Facial Expression Coefficient Prediction From Partial Face Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Virtual reality devices often collect only partial facial information, leading to inaccuracies in simulating complete facial expressions due to the limitations of existing algorithms in generating supervisory signals from incomplete facial data.
Innovation Solution
A method and apparatus for facial expression simulation that involves collecting local facial images, generating expression coefficients based on their position in an image sequence, and using an expression coefficient prediction model to accurately predict corresponding coefficients, enabling accurate facial expression simulation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If VR devices collect only partial facial information, then device complexity and cost are reduced, but facial expression simulation accuracy deteriorates
Solution Approach 1:
The patent introduces an expression coefficient prediction model as an intermediary component that bridges the gap between partial facial images and complete expression coefficients. The model takes local facial images (eyes, eyebrows, mouth) as input and predicts the corresponding expression coefficients, effectively mediating between the limited data from VR devices and the requirements for accurate facial expression simulation.
Solution Approach 2:
The patent transforms the problem from collecting complete facial images to collecting local facial images and predicting expression coefficients. By changing the parameter space from spatial coverage (complete face vs. local regions) to coefficient prediction, the system achieves accurate expression simulation with minimal facial data collection.
2Measurement precision
If complete facial information is collected to simulate accurate expressions, then simulation accuracy is improved, but data collection requirements and processing complexity increase
Solution Approach 1:
The patent extracts only the necessary local facial regions (eyes, eyebrows, mouth) from the complete face and uses these extracted regions to predict expression coefficients. This extraction approach eliminates the need to collect and process complete facial images while maintaining expression simulation accuracy, as the local regions contain sufficient information for coefficient prediction.
Solution Approach 2:
The patent segments the facial expression into local regions (eyes, eyebrows, mouth) and processes them independently. Each local region is analyzed to predict its corresponding expression coefficients, allowing the system to achieve complete expression simulation by combining results from segmented regions rather than processing the entire face at once.
Data Source
AI summary
The present disclosure provides a facial expression simulation method and apparatus, a device, and a storage medium. The method comprises: collecting a local facial image to be processed of a target object, and generating an expression coefficient corresponding to the local facial image to be processed, wherein the local facial image to be processed belongs to an expression image sequence, and the expression coefficient is determined on the basis of the position of the local facial image to be processed in the expression image sequence; and simulating a facial expression of the target object according to the expression coefficient.