DC Field | Value | Language |
---|---|---|
dc.contributor.author | Chung, Joon Son | ko |
dc.contributor.author | Zisserman, Andrew | ko |
dc.date.accessioned | 2021-11-27T06:40:34Z | - |
dc.date.available | 2021-11-27T06:40:34Z | - |
dc.date.created | 2021-11-26 | - |
dc.date.created | 2021-11-26 | - |
dc.date.created | 2021-11-26 | - |
dc.date.issued | 2018-08 | - |
dc.identifier.citation | COMPUTER VISION AND IMAGE UNDERSTANDING, v.173, pp.76 - 85 | - |
dc.identifier.issn | 1077-3142 | - |
dc.identifier.uri | http://hdl.handle.net/10203/289581 | - |
dc.description.abstract | Our aim is to recognise the words being spoken by a talking face, given only the video but not the audio. Existing works in this area have focussed on trying to recognise a small number of utterances in controlled environments (e.g. digits and alphabets), partially due to the shortage of suitable datasets. We make three novel contributions: first, we develop a pipeline for fully automated data collection from TV broadcasts. With this we have generated a dataset with over a million word instances, spoken by over a thousand different people; second, we develop a two-stream convolutional neural network that learns a joint embedding between the sound and the mouth motions from unlabelled data. We apply this network to the tasks of audio-to-video synchronisation and active speaker detection; third, we train convolutional and recurrent networks that are able to effectively learn and recognize hundreds of words from this large-scale dataset. In lip reading and in speaker detection, we demonstrate results that exceed the current state-of-the-art on public benchmark datasets. | - |
dc.language | English | - |
dc.publisher | ACADEMIC PRESS INC ELSEVIER SCIENCE | - |
dc.title | Learning to lip read words by watching videos | - |
dc.type | Article | - |
dc.identifier.wosid | 000454184600009 | - |
dc.identifier.scopusid | 2-s2.0-85044661381 | - |
dc.type.rims | ART | - |
dc.citation.volume | 173 | - |
dc.citation.beginningpage | 76 | - |
dc.citation.endingpage | 85 | - |
dc.citation.publicationname | COMPUTER VISION AND IMAGE UNDERSTANDING | - |
dc.identifier.doi | 10.1016/j.cviu.2018.02.001 | - |
dc.contributor.localauthor | Chung, Joon Son | - |
dc.contributor.nonIdAuthor | Zisserman, Andrew | - |
dc.description.isOpenAccess | N | - |
dc.type.journalArticle | Article | - |
dc.subject.keywordAuthor | Lip reading | - |
dc.subject.keywordAuthor | Lip synchronisation | - |
dc.subject.keywordAuthor | Active speaker detection | - |
dc.subject.keywordAuthor | Large vocabulary | - |
dc.subject.keywordAuthor | Dataset | - |
dc.subject.keywordPlus | SPEECH | - |
dc.subject.keywordPlus | EXTRACTION | - |
dc.subject.keywordPlus | FEATURES | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.