MonocularRGB_2D_Handjoints_MVA19 icon indicating copy to clipboard operation
MonocularRGB_2D_Handjoints_MVA19 copied to clipboard

Published 20 hours ago • verified publisher icon FORTH-ModelBasedTracker
→

Metadata

Accurate Hand Keypoint Localization on Mobile Devices

Accurate Hand Keypoint Localization on Mobile Devices

overview

Use the pretrained network of the paper "Accurate Hand Keypoints Localization on Mobile Devices" to extract hand joint heatmaps from RGB hand images. The network is light enought (based on a mobilenet2 backend) to work on mobile phones.

Download the paper from here.

supplementary

Requirements

Tested on Python 3.x. Required packages

  • numpy
  • tensorflow

Usage

Clone the repo and run joint_demo.py.

← Metadata

65
Stars
12
Forks
Watchers

Owner

verified publisher icon FORTH-ModelBasedTracker

Metadata

Accurate Hand Keypoint Localization on Mobile Devices

Back