Read and Attend: Temporal Localisation in Sign Language Videos - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

Read and Attend: Temporal Localisation in Sign Language Videos

Résumé

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.

Dates et versions

hal-03513396 , version 1 (05-01-2022)

Identifiants

Citer

Gül Varol, Liliane Momeni, Samuel Albanie, Triantafyllos Afouras, Andrew Zisserman. Read and Attend: Temporal Localisation in Sign Language Videos. 2021 IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2021), Jun 2021, Nashville, TN, United States. ⟨10.1109/CVPR46437.2021.01658⟩. ⟨hal-03513396⟩
20 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More