# A Novel Contactless Human Machine Interface based on Machine Learning

**Authors:** Frederic Magoules, Qinmeng Zou

arXiv: 1907.04390 · 2019-07-12

## TL;DR

This paper presents a contactless human-machine interface using computer vision and machine learning, allowing users to control computers with simple hand gestures via a standard camera, eliminating traditional input devices.

## Contribution

It introduces a novel framework that leverages simple camera input and machine learning for gesture-based control, simplifying human-computer interaction.

## Key findings

- Uses only a standard camera for interaction
- Employs machine learning for gesture detection and tracking
- Enables control of virtual interfaces with simple gestures

## Abstract

This paper describes a global framework that enables contactless human machine interaction using computer vision and machine learning techniques. The main originality of our framework is that only a very simple image acquisition device, as a computer camera, is sufficient to establish a rich human machine interaction as traditional devices such as mouse or keyboard. This framework is based on well known computer vision techniques and efficient machine learning techniques are used to detect and track user hand gestures so the end user can control his computer using virtual interfaces with very simple gestures.

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/1907.04390/full.md

## Figures

13 figures with captions in the complete paper: https://tomesphere.com/paper/1907.04390/full.md

## References

24 references — full list in the complete paper: https://tomesphere.com/paper/1907.04390/full.md

---
Source: https://tomesphere.com/paper/1907.04390