Table 1 Summary of the datasets. The mean quantity of video instances per class is indicated in column “Mean”.

From: A deep learning-based method combines manual and non-manual features for sign language recognition

Dataset

Language

Sensor

Gloss

Videos

Mean

Signers

WLASL300

ASL

RGB

300

5117

17.1

109

AUTSL

TSL

RGB+D

226

38,336

169.6

43

Back to article page