How2Sign: A Large-scale Multimodal Dataset for Continuous American Sign Language
This dataset addresses a bottleneck for researchers in sign language recognition, translation, and production, enabling more robust models, though it is incremental as it builds on existing data collection efforts.
The authors tackled the lack of large annotated datasets for sign language research by introducing How2Sign, a multimodal dataset with over 80 hours of American Sign Language videos and corresponding modalities like speech and transcripts, and they showed that synthesized videos from this dataset can be understood by ASL signers.
One of the factors that have hindered progress in the areas of sign language recognition, translation, and production is the absence of large annotated datasets. Towards this end, we introduce How2Sign, a multimodal and multiview continuous American Sign Language (ASL) dataset, consisting of a parallel corpus of more than 80 hours of sign language videos and a set of corresponding modalities including speech, English transcripts, and depth. A three-hour subset was further recorded in the Panoptic studio enabling detailed 3D pose estimation. To evaluate the potential of How2Sign for real-world impact, we conduct a study with ASL signers and show that synthesized videos using our dataset can indeed be understood. The study further gives insights on challenges that computer vision should address in order to make progress in this field. Dataset website: http://how2sign.github.io/