Human Posture Detection Using Deep Learning

Abstract
Human Posture detection using deep learning is a widely applied approach used to detect the movement of a person’s limb. The purpose of this project is to develop a generic method for detecting global human body position in video sequences. We have used 5 datasets named UP Fall, URFD, NUCLA, MCF, and KARD. We implemented these datasets on 7 different architectures, ResNet 50, ResNet 101, ResNet 152, VGG 16, VGG 19, DenseNet 121, and Inception V3. Best-obtained results were on ResNet 50 with an average accuracy of 98%. We are working on the implementation of the project.
Description
Keywords
Citation
Collections