Skip to content
master
Switch branches/tags
Go to file
Code

Latest commit

Hi! Here's the changes I made to make the hand_model.ipynb work: 1- The /model/ directory was added to the path for metalwhale repo, 2- The file name for Google Mediapipe was changed from hand_landmark_3d.tflite to hand_landmark.tflite , and 3- I changed the --referer argument from the file GitHub address to the raw file, because the original address gave me the following error:
""""
.
.
.
./models/palm_detec 100%[===================>]   6.74M  2.78MB/s    in 2.4s    

2020-08-28 12:43:36 (2.78 MB/s) - ‘./models/palm_detection.tflite’ saved [7069988/7069988]

palm_detection done

libc++abi.dylib: terminating with uncaught exception of type std::runtime_error: Couldn't close file """
d115745

Git stats

Files

Permalink
Failed to load latest commit information.

README.md

Hand Tracker

Binder

Simple Python wrapper for Google's Mediapipe Hand Tracking pipeline.

Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.

Requirements

These are required to use the HandTracker module

numpy
opencv
tensorflow

To run the Jupyter demo you'll also need jupyter and matplotlib

Setup

Download models using load_models.sh, or load them manually from metalwhale's repo and put them inside models dir.

Anchors

To get the SSD anchors I've written a C++ program that executes the SsdAnchorsCalculator::GenerateAnchors function from this calculator. As there's no reason to modify provided anchors I do not include it into the repository, but you can find the script here

Acknowledgments

This work is a study of models developed by Google and distributed as a part of the Mediapipe framework.
Big thanks to @metalwhale for removing cusom operation dependencies.

About

Minimal Python interface for Google's Mediapipe HandTracking pipeline

Resources

License

Releases

No releases published

Packages

No packages published