Thriving 
PY This repository contains utilities for loading data from Meta's datasets with hand pose annotations, computing hand tracking metrics, and visualising results of a hand tracking method.
HEALTH ? 73 / Stable
Thriving PY // readme
A toolkit for egocentric hand tracking research
Check out the Multiview Egocentric Hand Tracking Challenge!!
This repository contains the following tools for hand tracking research:
- APIs for loading data from the UmeTrack and HOT3D datasets
- Computation of metrics used in the Hand Tracking Challenge
- Data visualization helpers
Datasets
The datasets are provided using the WebDataset format. The file structure of each sequence is as follows:
├─ train
│ ├─ subject_000_separate_hand_000000.tar
│ ├─ subject_000_hand_hand_000001.tar
│ ├─ ...
│ ├─ subject_049_hand_hand_000099.tar
├─ pose_test
│ ├─ subject_050_separate_hand_000100.tar
│ ├─ subject_050_hand_hand_000101.tar
│ ├─ ...
│ ├─ subject_079_hand_hand_000199.tar
├─ shape_test
│ ├─ subject_080_separate_hand_000300.tar
│ ├─ subject_080_hand_hand_000301.tar
│ ├─ ...
│ ├─ subject_099_hand_hand_000399.tar
Each tar file is assumed to contain 2~4 synchronized monochrome streams plus an optional RGB stream. The image files are suffixed with the image stream ID which can be used to…