The dataset viewer is not available because its heuristics could not detect any supported data files. You can try uploading some data files, or configuring the data files location manually.
Dataset Card for easy-to-read translation
-Table of Contents -Dataset Description -Dataset Summary -Supported Tasks and Leaderboards -Languages -Dataset Structure -Data Instances -Data Fields -Dataset Creation -Source Data -Annotations
Dataset Summary
A corpus with a subset of 20,920 videos of 30 fingerspelled letters of the LSE (which comes from Lengua de Signos Española) alphabet and five seleccted signs ("Good morning", "Good afternoon", "Good night", "Thank you" and "Claps"). This dataset was compiled with the aim of creating a multi-person dataset to reduce the language barrier for people with hearing impairments.
Supported Tasks and Leaderboards
Video Classification
Languages
Spanish and Spanish Sign Language.
Dataset Structure
Data Instances
For each instance, there is a string for
The dataset is composed of 20,920 videos in .avi and .mp4 formats and a txt for each video with the skeleton data.
Data Fields
The dataset contains 18 folders containing all the videos. The most important information about each video is provided in its name. Each video name contains the background it was recorded on, the camera, the viewing angle, video type (RGB, Depth, or IR), gesture, hand used, participant ID, and gesture repetition number. Moreover, there is a folder with all the txt files.
Source Data
The dataset was taken with 4 different cameras: 2x Logitech cameras at 640x480 resolution and 30 FPS, 2x Microsoft cameras at 640x480 resolution and 30 FPS, 1x FEMTO that captures 3 different types of video; RGB at 1920x1080 resolution and 30 FPS, Depth at 640x576 resolution at 30 FPS and Infrared at 640x576 resolution at 30 FPS and 1x Realsense that captures 2 different types of video; RGB at 640x480 resolution at 30 FPS and Depth at 640x480 resolution at 30 FPS.
Annotations
The dataset contains a total of 20,920 videos divided between the letters of the LSE finger alphabet and five simple actions (Aplausos, Buenos Días, Buenas Tardes, Buenas Noches and Gracias). The minimum number of videos per class for letters is XXX and around XXX for the rest of the letters. The maximum number of videos for a letter is XXX and for actions, a minimum of XXX per action (for the action Gracias) and a maximum of XXX (for the action Buenos Días).
Licensing
Alguna gratuita y de código abierto tipo GPL
Citation information
Aquí poner la referencia para poder citarlo en BibTex
- Downloads last month
- 7