Autoencoder Facial Normalization for Cross-Group Expression Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing facial expression recognition technologies struggle with generalization across different groups of people due to variations in facial appearance, demographics, and data collection settings, leading to performance gaps and biases.
Innovation Solution
A self-supervised denoising autoencoder is used to transfer facial expressions onto a common facial template, separating the learning process into phases that reduce individual differences while preserving dynamic characteristics, utilizing a single encoder and multiple decoders trained on reconstruction losses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional face normalization techniques are used, then performance is good on input data with similar characteristics to training data, but performance degrades on data with diverse demographic characteristics
Solution Approach 1:
The patent introduces a normalization autoencoder as an intermediary component between the input face image and the FAU classification model. This autoencoder learns to map diverse facial appearances to a standardized representation space, serving as a mediator that bridges the gap between different demographic groups and the classifier, thereby improving generalization without sacrificing reliability on any specific group
Solution Approach 2:
The patent segments the facial image processing into two distinct stages: first, the normalization autoencoder processes the raw face image to extract and normalize facial features, separating the demographic variation from the expressive content; second, the FAU classifier processes the normalized features. This segmentation allows each component to specialize, with the autoencoder handling demographic diversity and the classifier focusing on expression recognition
2Measurement precision
If the autoencoder is trained to preserve all individual characteristics, then individual fidelity is maintained, but performance on diverse groups deteriorates
Solution Approach 1:
The patent transforms the training objective parameters of the autoencoder by using reconstruction loss as the optimization target. This parameter change guides the autoencoder to learn a representation that preserves essential facial features while implicitly normalizing demographic variations, achieving a balance between individual fidelity and cross-group adaptability through the mathematical formulation of the loss function
Data Source
AI summary
Generally discussed herein are devices, systems, and methods for. A method can include obtaining a normalizing autoencoder, the normalizing autoencoder trained based on first data samples of a template person and second data samples of a variety of people, normalizing, by the normalizing autoencoder, an input data sample by combining dynamic characteristics of a person in the input data sample with static characteristics in the first data samples, to generate normalized data, and providing the normalized data as input to a classifier model to classify the input data based on the dynamic characteristics of the input data and the static characteristics of the first data samples.