Human pose estimation based on data-driven Monte Carlo hidden Markov models

TL;DRAbstract

Estimating human poses from 2D images or video sequences can provide the moving trajectories of the body joints for the high level processing, human activity recognition, which is applicable in surveillance, human-computer interaction and clinical and sport analysis. The underlying concept of a human pose estimation system is to find the best hypothesis or human model that fits the representation of the body pose built from the observed image. Hence such a system normally consists of three parts, 1. human detection for finding the global location of the human inside the image;

Chat with Paper

AI Agents for this Paper