https://github.com/epic-kitchens/epic-kitchens-55-lib

:coffee: EPIC-KITCHENS-55 dataset python library
https://github.com/epic-kitchens/epic-kitchens-55-lib

action-recogntion dataset egocentric-action-recognition epic-kitchens helper library python pytorch tensorflow

Last synced: 7 months ago
JSON representation

:coffee: EPIC-KITCHENS-55 dataset python library

• Host: GitHub
• URL: https://github.com/epic-kitchens/epic-kitchens-55-lib
• Owner: epic-kitchens
• Created: 2018-04-10T15:55:10.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2022-06-21T21:18:23.000Z (about 4 years ago)
• Last Synced: 2025-07-07T04:05:23.993Z (about 1 year ago)
• Topics: action-recogntion, dataset, egocentric-action-recognition, epic-kitchens, helper, library, python, pytorch, tensorflow
• Language: Python
• Homepage: https://epic-kitchens.readthedocs.io/en/latest/index.html
• Size: 156 KB
• Stars: 31
• Watchers: 3
• Forks: 6
• Open Issues: 10
• Metadata Files:
  • Readme: README.md
  • Changelog: CHANGELOG.md

Awesome Lists containing this project

README

          
# EPIC-KITCHENS-55 python library

[![CircleCI status badge](https://img.shields.io/circleci/project/github/epic-kitchens/epic-lib/master.svg)](https://circleci.com/gh/epic-kitchens/epic-lib)

[![codecov](https://codecov.io/gh/epic-kitchens/epic-lib/branch/master/graph/badge.svg)](https://codecov.io/gh/epic-kitchens/epic-lib)

[![Documentation Status](https://readthedocs.org/projects/epic-kitchens/badge/?version=stable)](http://epic-kitchens.readthedocs.io/en/stable/?badge=stable)

[![PyPI](https://img.shields.io/pypi/v/epic-kitchens.svg)](https://pypi.org/project/epic-kitchens/#description)

[![PyPI - Python Version](https://img.shields.io/pypi/pyversions/epic-kitchens.svg)](https://pypi.org/project/epic-kitchens/)

> A library for easily integrating the EPIC-KITCHENS-55 egocentric dataset in your

> experiments

## Install

```console

$ pip install epic-kitchens

```

## Details

* Segmentation scripts for splitting raw video frames/flow into action segments

* [GulpIO adapter](https://github.com/TwentyBN/GulpIO#loading-data) for ingesting and reading the

  dataset. This works particularly well for PyTorch and Tensorflow models.

* Dataset classes for loading and augmenting data

* Utilities for converting between frame indices for RGB and flow.

See the [CHANGELOG](CHANGELOG.md) for release notes

**WARNING**: If you have used v1.4.0 or below to gulp flow, please delete the

data and regulp due to an issue in reading files from disk, the gulped frames will

be unordered instead of ordered by frame index (i.e. lexicographically).

Check out the documentation on [Read the

docs](https://epic-kitchens.readthedocs.io/en/stable/index.html)