Gait Sequence

Gait sequence analysis focuses on understanding and characterizing the temporal and spatial patterns of human locomotion, aiming to extract meaningful information for various applications. Current research emphasizes developing robust methods for gait cycle segmentation and feature extraction using diverse data sources (e.g., inertial measurement units, video, 3D motion capture), often employing deep learning architectures like LSTMs, convolutional neural networks, and graph convolutional networks, along with techniques like optimal transport and Koopman operator theory. These advancements have significant implications for clinical gait assessment (e.g., diagnosing neurological disorders), human identification (e.g., security and surveillance), and robotics (e.g., improving legged robot locomotion).