Skip to content

Latest commit

 

History

History
12 lines (9 loc) · 507 Bytes

README.md

File metadata and controls

12 lines (9 loc) · 507 Bytes

AI-virtual-mouse-mediapipe

Hand Tracking: we are using two module at the backhand

1. palm detection:

	it bascially work on full image of the hand and provides the crop image of only palm

2. Hand Landmarks:

it module find 21 different landmarks on the cropped image send by the palm detection..

image

To train this hand landmark the manually annoted 30 thousand images of different hand.