Pose estimation method, pose estimation device and computer system

Abstract

The invention relates to a pose estimation method, a pose estimation device and a computer system. The pose estimation method comprises the steps of extracting features corresponding to body parts of an object from an input image so as to generate a feature graph; updating the feature graph based on a predetermined graph model of the object; building a body part score graph according to the updated feature graph; determining the position of each body part of the object in the input image through the body part score graph; and estimating the pose of the object according to the determined positions.

Description

technical field [0001] The present application generally relates to the field of computer technology, specifically to computer vision, and more specifically to a pose estimation method, device and computer system. Background technique [0002] There are many existing works trying to apply the graphical model to the deep learning neural network. But these attempts are relatively simple, such as directly using the graphical model as a post-processing to improve the accuracy of the results, or jointly optimizing the graphical model and neural network of the prediction layer. Since the information of the prediction layer of the neural network is relatively small, generally speaking, there is only one value, representing whether it is a certain key point of the body, or clustering is performed according to the two-dimensional rotation angle to obtain information containing two-dimensional angles. But a lot of visual information is lost, such as what kind of clothes the person we...

AI Technical Summary

Problems solved by technology

This information is very helpful for the prediction of key points, but it has been compressed and lost in the prediction layer

Images