BSL-1K: Scaling up co-articulated sign language recognition using mouthing cues
This work addresses the data scarcity problem for sign language recognition researchers, enabling more accurate models for applications like translation, though it is incremental in scaling up existing methods.
The authors tackled the lack of training data for automated sign language recognition by introducing BSL-1K, a dataset of 1,000 British Sign Language signs collected from 1,000 hours of video using mouthing cues and weakly-aligned subtitles. They showed that models trained on this dataset exceed state-of-the-art performance on MSASL and WLASL benchmarks.
Recent progress in fine-grained gesture and action classification, and machine translation, point to the possibility of automated sign language recognition becoming a reality. A key stumbling block in making progress towards this goal is a lack of appropriate training data, stemming from the high complexity of sign annotation and a limited supply of qualified annotators. In this work, we introduce a new scalable approach to data collection for sign recognition in continuous videos. We make use of weakly-aligned subtitles for broadcast footage together with a keyword spotting method to automatically localise sign-instances for a vocabulary of 1,000 signs in 1,000 hours of video. We make the following contributions: (1) We show how to use mouthing cues from signers to obtain high-quality annotations from video data - the result is the BSL-1K dataset, a collection of British Sign Language (BSL) signs of unprecedented scale; (2) We show that we can use BSL-1K to train strong sign recognition models for co-articulated signs in BSL and that these models additionally form excellent pretraining for other sign languages and benchmarks - we exceed the state of the art on both the MSASL and WLASL benchmarks. Finally, (3) we propose new large-scale evaluation sets for the tasks of sign recognition and sign spotting and provide baselines which we hope will serve to stimulate research in this area.