Read and Attend: Temporal Localisation in Sign Language Videos
This addresses sign language recognition for accessibility applications, though it is incremental as it builds on existing Transformer and weak supervision methods.
The paper tackles the problem of localizing sign instances in continuous sign language videos by training a Transformer model on weakly-aligned subtitles, enabling automatic annotation of hundreds of thousands of signs and outperforming prior state-of-the-art on the BSL-1K benchmark.
The objective of this work is to annotate sign instances across a broad vocabulary in continuous sign language. We train a Transformer model to ingest a continuous signing stream and output a sequence of written tokens on a large-scale collection of signing footage with weakly-aligned subtitles. We show that through this training it acquires the ability to attend to a large vocabulary of sign instances in the input sequence, enabling their localisation. Our contributions are as follows: (1) we demonstrate the ability to leverage large quantities of continuous signing videos with weakly-aligned subtitles to localise signs in continuous sign language; (2) we employ the learned attention to automatically generate hundreds of thousands of annotations for a large sign vocabulary; (3) we collect a set of 37K manually verified sign instances across a vocabulary of 950 sign classes to support our study of sign language recognition; (4) by training on the newly annotated data from our method, we outperform the prior state of the art on the BSL-1K sign language recognition benchmark.