GSLL Dataset

Greek Sign Language Lemmas Dataset (GSLL Dataset)

Greek Sign Language Lemmas Dataset (GSLL Dataset):

This is a Greek sign language dataset for isolated sign language recognition. This dataset contains 347 different signs/classes signed by two signers: Kostas and Olga (male and female). Each sign is repeated from 5 times to 17. These 347 classes are recorded through a total of 3,464 videos containing 161,050 frames. The dataset is 42 GB. You can download the dataset here.

The dataset is organized as follows:

| |__1-1
| |__1-2
| |__ …
| |__1-1
| |__1-2
| |__…

where 1-1,1-2,… are the classes folders containing repetitions of each sign. For example, the folder /kostas/1-1/ refers to the sign “a little”, while the folder /olga/5-6/ refers to the sign “cat”.

The meaning of the signs can be found in the GSLL_LEMMAS.xsl, using the F-G columns for the enumeration and the column B for the sign concept. You can download GSLL_LEMMAS.xsl here.

Selection 049

Table: Statistics for the Greek Sign Language Lemmas Dataset and its respective subsets. Indicative suggested splitting in train, dev and test set used in the experiments of [1].


Figure: Characteristic frames the Greek Sign Language Lemmas Dataset from both signers.

Reference to:

In case you use the GSLL dataset in your work, please cite:

[1] A. Kratimenos, G. Pavlakos and P. Maragos, “Independent Sign Language Recognition with 3D Body, Hands, and Face Reconstruction”, in Proc. IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2021.

[2] S. Theodorakis, V. Pitsikalis, and P. Maragos, “Dynamic-Static Unsupervised Sequentiality, Statistical Subunits and Lexicon for Sign Language Recognition”, Image and Vision Computing, vol.32, no.8, pp.533–549, Aug. 2014.

[3] E. Efthimiou, S-E. Fotinea, C. Vogler, T. Hanke, J. Glauert, R. Bowden, A. Braffort, C. Collet, P. Maragos, and J. Segouat, “Sign Language Recognition, Generation, and Modelling: A Research Effort with Applications in Deaf Communication”, Proc. 5th Conf. on Universal Access in Human-Computer Interaction (jointly with HCI International 2009), Springer LNCS vol.5614, pp. 21-30, 2009.


The development of the GSLL dataset was supported by the EU research project Dicta-Sign with grant FP7-ICT-3-231135.


For more information or details regarding the dataset contact:

Agelos Kratimenos
Research Assistant, IRAL-CVSP Lab
National Technical University of Athens



Prof. Petros Maragos
Director IRAL-CVSP Lab
National Technical University of Athens