GENERATION OF SYNTHETIC MOVEMENTS BY GENERATIVE ADVERSARIAL NETWORKS FOR IMPROVING THE RECOGNITION OF RARE ACTIONS

Abstract

Action recognition systems suffer significantly when dealing with rare or underepresented action classes. When training data are scarce, these systems show poor generalization and develop biased behavior that favors common actions. This study presents a novel GAN-based framework designed specifically to create realistic motion sequences for rare action classes in video datasets. The methodology combines temporal generative adversarial networks with motion-aware discriminators. This combination produces high-quality synthetic action sequences that maintain both the visual appearance and timing patterns of real human movement. The approach uses a specialized architecture with three-dimensional convolutional layers, temporal attention mechanisms, and physical constraints. These components work together to ensure the generated motions look and feel realistic from a biomechanical perspective. The framework operates through a two-stage training process. First, a motion encoder analyzes existing rare action samples to extract key movement patterns. Then, a conditional GAN generates new motion sequences while keeping the essential characteristics that make each action unique.           Testing on standard benchmarks (UCF-101, HMDB-51, and NTU RGB+D) shows substantial improvements in the rare action recognition accuracy. The synthetic data augmentation approach delivers an average performance boost of 23.7% for actions with fewer than 50 training samples and 31.2% for actions with fewer than 20 samples across all datasets. Complex actions show even more dramatic improvements, with medical procedures improving by 42.8% and specialized sports movements by 38.4%. Quality assessment using Fréchet Video Distance (FVD) metrics and human perception studies confirms that generated motions are visually indistinguishable from real sequences. The synthetic movements maintain the characteristic timing patterns that define each action type. These findings demonstrate that AI-driven synthetic data generation can effectively solve class imbalance problems in action recognition, leading to more robust and fair model performance across all action categories.