Synthetic Data Generation for Emotional Depth Faces: Optimizing Conditional DCGANs via Genetic Algorithms in the Latent Space and Stabilizing Training with Knowledge Distillation
This addresses a data scarcity problem in affective computing for researchers and developers, though it appears incremental as it builds on existing GAN techniques with optimizations.
The paper tackles the lack of high-quality, diverse depth facial datasets for emotional recognition by proposing a synthetic data generation framework using an optimized GAN with Knowledge Distillation and Genetic Algorithms, achieving 94-96% classification accuracy and consistent improvements in metrics like FID and SSIM over state-of-the-art methods.
Affective computing faces a major challenge: the lack of high-quality, diverse depth facial datasets for recognizing subtle emotional expressions. We propose a framework for synthetic depth face generation using an optimized GAN with Knowledge Distillation (EMA teacher models) to stabilize training, improve quality, and prevent mode collapse. We also apply Genetic Algorithms to evolve GAN latent vectors based on image statistics, boosting diversity and visual quality for target emotions. The approach outperforms GAN, VAE, GMM, and KDE in both diversity and quality. For classification, we extract and concatenate LBP, HOG, Sobel edge, and intensity histogram features, achieving 94% and 96% accuracy with XGBoost. Evaluation using FID, IS, SSIM, and PSNR shows consistent improvement over state-of-the-art methods.