Florence 4D Facial Expression Dataset

Florence 4D Facial Expression Dataset是由佛罗伦萨大学媒体集成与通信中心创建的大型数据集，包含动态的3D面部表情序列。该数据集结合了真实与合成身份，展示了前所未有的4D面部表情多样性，包括从经典的中性到顶峰过渡，以及表情之间的过渡。数据集包含95个身份，总计205,200个序列，每个序列都具有随机化的速度以增强真实感。该数据集旨在推动4D面部表情分析的研究，特别是在动态表情识别和生成领域。

The Florence 4D Facial Expression Dataset is a large-scale dataset created by the Media Integration and Communication Center at the University of Florence. It contains dynamic 3D facial expression sequences, combining real and synthetic identities to showcase unprecedented diversity in 4D facial expressions, including transitions from neutral expressions to their peak intensity as well as transitions between distinct facial expressions. The dataset includes 95 unique identities, totaling 205,200 sequences, each featuring randomized playback speeds to enhance realism. This dataset aims to advance research in 4D facial expression analysis, particularly in the domains of dynamic facial expression recognition and synthesis.