Context-aware human motion generation: A comprehensive survey
Boyuan Cheng, Zixuan Zhou, Yajie Deng, Siyao Du, Zhengyezi Wang, Wen Yi, Yixuan Zhang, Shang Ni, Yanzhe Kong, L Y Xie, Xiaosong Yang · Design and Artificial Intelligence · 2025
Context-aware human motion generation involves synthesizing human poses and movements that align with specific input conditions. This technique plays a crucial role in various domains. Despite its potential, generating realistic and physically plausible human motion remains a challenging task because of the complexity of human dynamics, high-quality dataset requirements, and the need for temporal consistency and condition alignment. This paper provides a comprehensive review of the current state of conditioned human motion generation, focusing on several key aspects: motion representation, generative models , datasets, methodologies, evaluation metrics, and persistent challenges. In addition, this review summarizes current advancements and suggests future research directions to enhance context-aware human motion generation technologies.