Slovo: Russian Sign Language Dataset

Alexander Kapitanov, Karina Kvanchiani, Alexander Nagaev, Elizaveta Petrova
5/23/2023
cs.CV

No official code found. Best implementation: hukenovs/slovo (90 stars, last commit 2024-08-22). 1 more below.

Abstract

One of the main challenges of the sign language recognition task is the difficulty of collecting a suitable dataset due to the gap between hard-of-hearing and hearing societies. In addition, the sign language in each country differs significantly, which obliges the creation of new data for each of them. This paper presents the Russian Sign Language (RSL) video dataset Slovo, produced using crowdsourcing platforms. The dataset contains 20,000 FullHD recordings, divided into 1,000 classes of isolated RSL gestures received by 194 signers. We also provide the entire dataset creation pipeline, from data collection to video annotation, with the following demo application. Several neural networks are trained and evaluated on the Slovo to demonstrate its teaching ability. Proposed data and pre-trained models are publicly available.

View on arXivView PDF

Code Implementations(2)

hukenovs/slovoCommunity100%

Slovo: Russian Sign Language Dataset and Models

9010PythonApr 2, 20232 years ago
datasetdeep-learninggesture-recognitiongestures-classificationglosses+4 more
Apache-2.0

implementation of a custom model for Russian SLR (Sign Language Recognition) on the dataset "SLOVO" https://arxiv.org/abs/2305.14527

110Python, DockerfileJun 22, 20256 months agoApache-2.0

Similar Papers

Cite this paper

@article{kapitanov2023slovo,
  title  = {Slovo: Russian Sign Language Dataset},
  author = {Alexander Kapitanov and Karina Kvanchiani and Alexander Nagaev and Elizaveta Petrova},
  year   = {2023},
  eprint = {2305.14527},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2305.14527v3}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2305.14527.svg)](https://codeofpaper.com/paper/2305.14527)

Discussion