HUMAN BEHAVIOUR THROUGH COMPUTER VISION
This is the repository concerning my research about the detection of human behaviour using computer vision. My approach is based on human skeleton.
The goal of this research is to determined whether a pose-based approach is a good solution to implement human behaviour detection. The assertion is that an machine learning algorithm will be more efficient with numeric data, instead of direct image recognition.
By detecting the human skeleton, the algorithm will be able to work with "bones" coordinates and orientation, which should increase the detection accuracy, and IA practice.
Author : Anatole de Chauveron
Keywords : Computer Vision, OpenCV, OpenPose, Human Behaviour, Human Poses, Machine Learning, Form detection
Begin date : 1 november 2021
Last update : 03 january 2022
Licence : All rights reserved
Informations
For questions or futher informations about this subject contact me at adechauveron@gmail.com.