MULTI-VIEW 3D HUMAN POSE ESTIMATION METHOD AND RELATED APPARATUS

摘要

A multi-view 3D human pose estimation method and a related apparatus. The method comprises: a server obtains a 3D human pose at a first moment in a continuous time period until 3D human poses at all moments in the continuous time period are obtained, the continuous time period comprising multiple moments, and the first moment being any moment in the continuous time period (101); the server inputs the 3D human poses at all the moments in the continuous time period into a target LSTM network model to obtain an estimated 3D human pose at the next moment in the continuous time period (102). By means of the method, the precision of the estimated 3D human pose can be improved.