Edge Impulse Experts / Hand Gestures Public

Hand Gestures

In this simulation, we will create four gesture classes: neutral (fist), five (open hand), peace (V-sign), and good (thumbs up). The bounding box detection will provide class data along with the object’s x, y, w, and h values, which we will use as real-time input to replace a keyboard or joystick in the classic game we’re developing. The trained model will then be deployed to a Raspberry Pi, integrated into our Python code that uses the PyGame library for easy development and rendering of the game on an LCD display.

Object detection

About this project

This public Edge Impulse project does not have a README yet. Clone this project to add new data or retrain this project, or to deploy this project to a device.
hand00098
hand00058
hand00075
hand00069
hand00050
hand00102
hand00051
hand00022

Run this model

On any device

Dataset summary

Data collected
107 items
Labels
five, good, neut, peace

Project info

Project ID 766818
License 3-Clause BSD
No. of views 4,201
No. of clones 10