Data processing method and training method and device of motion prediction model

Abstract

The invention discloses a data processing method and a motion prediction model training method and device.The data processing method comprises the steps that continuous action frames of at least one captured target object are obtained, and two or more motion capture mark points of the target object are marked in each action frame of the continuous action frames; according to the continuous action frames, obtaining initial feature data of the motion capture mark points in the continuous action frames; inputting the initial feature data into a motion prediction model, obtaining prediction feature data with an action inter-frame relationship and an action intra-frame relationship through a denoising network of the motion prediction model, and obtaining an action sequence of at least one motion capture mark point of the target object in continuous action frames by the motion prediction model according to the prediction feature data; wherein the action intra-frame relationship comprises an association relationship among different action capture mark points in each action frame, and the action inter-frame relationship comprises an association relationship among same action capture mark points in each action frame.

Description

technical field [0001] The present application relates to the field of data processing, and in particular to a data processing method, a training method and apparatus for a motion prediction model. Background technique [0002] Motion Capture (MoCap) is a technology that uses specific external equipment to record and restore the motion of moving objects. In recent years, with the development of related technologies, optical motion capture technology based on computer vision principles has become a commonly used motion capture method in the industry, and this technology is widely used in games, movies and other related motion production processes. [0003] In the specific application process, the optical motion capture technology uses multiple high-speed cameras with different perspectives to synchronously capture the movements of the actor, and calculates the three-dimensional coordinates of the marker points (maker) attached to different parts of the actor's body through th...

AI Technical Summary

Problems solved by technology

[0006] This application provides a data processing method to solve the problems of high cost, heavy workload and cumbersome process of denoising motion capture data existing in the prior art

Images