LSA-T: The First Continuous Argentinian Sign Language Dataset for Sign Language TranslationConference paper

Pedro Dal Bianco, Gastón Ríos, Franco Ronchetti, Facundo Quiroga, Oscar Stanchi, Waldo Hasperué, Alejandro Rosete

IBERAMIA 2022, Ibero-American Conference on Artificial Intelligence (LNCS, Springer)

The first continuous Argentinian Sign Language (LSA) dataset: 14,880 sentence-level videos from the CN Sordos YouTube channel, with Spanish labels and keypoints for each signer, a method to infer the active signer, a visualization tool and a baseline translation model.