SEEHEAR: signer diarisation and a new dataset
In this work, we propose a framework to collect a large-scale, diverse sign language dataset that can be used to train automatic sign language recognition models.The first contribution of this work is SDTrack, a generic method for signer tracking and diarisation in the wild. Our second contribution...
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Conference item |
| Language: | English |
| Published: |
IEEE
2021
|
| _version_ | 1797104049982537728 |
|---|---|
| author | Albanie, S Varol, G Momeni, L Afouras, T Brown, A Zhang, C Coto, E Camgoz, NC Saunders, B Dutta, A Fox, N Bowden, R Woll, B Zisserman, A |
| author_facet | Albanie, S Varol, G Momeni, L Afouras, T Brown, A Zhang, C Coto, E Camgoz, NC Saunders, B Dutta, A Fox, N Bowden, R Woll, B Zisserman, A |
| author_sort | Albanie, S |
| collection | OXFORD |
| description | In this work, we propose a framework to collect a large-scale, diverse sign language dataset that can be used to train automatic sign language recognition models.The first contribution of this work is SDTrack, a generic method for signer tracking and diarisation in the wild. Our second contribution is SeeHear, a dataset of 90 hours of British Sign Language (BSL) content featuring more than 1000 signers, and including interviews, monologues and debates. Using SDTrack, the SeeHear dataset is annotated with 35K active signing tracks, with corresponding signer identities and subtitles, and 40K automatically localised sign labels. As a third contribution, we provide benchmarks for signer diarisation and sign recognition on SeeHear. |
| first_indexed | 2024-03-07T06:28:24Z |
| format | Conference item |
| id | oxford-uuid:f51e3c20-b726-4988-8c8b-0439c0626f69 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T06:28:24Z |
| publishDate | 2021 |
| publisher | IEEE |
| record_format | dspace |
| spelling | oxford-uuid:f51e3c20-b726-4988-8c8b-0439c0626f692022-03-27T12:24:57ZSEEHEAR: signer diarisation and a new datasetConference itemhttp://purl.org/coar/resource_type/c_5794uuid:f51e3c20-b726-4988-8c8b-0439c0626f69EnglishSymplectic ElementsIEEE2021Albanie, SVarol, GMomeni, LAfouras, TBrown, AZhang, CCoto, ECamgoz, NCSaunders, BDutta, AFox, NBowden, RWoll, BZisserman, AIn this work, we propose a framework to collect a large-scale, diverse sign language dataset that can be used to train automatic sign language recognition models.The first contribution of this work is SDTrack, a generic method for signer tracking and diarisation in the wild. Our second contribution is SeeHear, a dataset of 90 hours of British Sign Language (BSL) content featuring more than 1000 signers, and including interviews, monologues and debates. Using SDTrack, the SeeHear dataset is annotated with 35K active signing tracks, with corresponding signer identities and subtitles, and 40K automatically localised sign labels. As a third contribution, we provide benchmarks for signer diarisation and sign recognition on SeeHear. |
| spellingShingle | Albanie, S Varol, G Momeni, L Afouras, T Brown, A Zhang, C Coto, E Camgoz, NC Saunders, B Dutta, A Fox, N Bowden, R Woll, B Zisserman, A SEEHEAR: signer diarisation and a new dataset |
| title | SEEHEAR: signer diarisation and a new dataset |
| title_full | SEEHEAR: signer diarisation and a new dataset |
| title_fullStr | SEEHEAR: signer diarisation and a new dataset |
| title_full_unstemmed | SEEHEAR: signer diarisation and a new dataset |
| title_short | SEEHEAR: signer diarisation and a new dataset |
| title_sort | seehear signer diarisation and a new dataset |
| work_keys_str_mv | AT albanies seehearsignerdiarisationandanewdataset AT varolg seehearsignerdiarisationandanewdataset AT momenil seehearsignerdiarisationandanewdataset AT afourast seehearsignerdiarisationandanewdataset AT browna seehearsignerdiarisationandanewdataset AT zhangc seehearsignerdiarisationandanewdataset AT cotoe seehearsignerdiarisationandanewdataset AT camgoznc seehearsignerdiarisationandanewdataset AT saundersb seehearsignerdiarisationandanewdataset AT duttaa seehearsignerdiarisationandanewdataset AT foxn seehearsignerdiarisationandanewdataset AT bowdenr seehearsignerdiarisationandanewdataset AT wollb seehearsignerdiarisationandanewdataset AT zissermana seehearsignerdiarisationandanewdataset |