Description

The NationalCSL-DP dataset is constructed by the Sichuan Province Key Laboratory of Philosophy and Social Science for Language Intelligence in Special Education. The goals to create the Chinese sign language dataset include: providing the world-wide researchers of SLR community a large vocabulary Chinese SL dataset for training and evaluating their algorithms; advancing the SOTA ISLR technologies aiming at practical applications especially for handling the issue of hand gesture occlusion. The NationalCSL-DP dataset covers 6707 standard Chinese Sign Language vocabularies in CNSL and totally contains 134140 sign videos recorded by 10 individuals (2 males and 8 females).

The description of dataset construction has been submitted to Scientific Data. A preprocessed dataset has been uploaded to https://figshare.com/account/articles/27261843. The uploaded dataset contains frames extracted from all sign language videos.

Considering that some researchers may wish to obtain the raw data, we place the raw data on our laboratory website for researchers to download.

Download

Signer 1: Front, Left

Signer 2: Front, Left

Signer 3: Front, Left

Signer 4: Front, Left

Signer 5: Front, Left

Signer 6: Front, Left

Signer 7: Front, Left

Signer 8: Front, Left

Signer 9: Front, Left

Signer 10: Front, Left