ECCV 2022 Sign Spotting Challenge: Dataset, Design and Results

Manuel Vázquez Enríquez*, José L.Alba Castro, Laura Docio Fernandez, Julio C.S. Jacques Junior, Sergio Escalera


Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review


The ECCV 2022 Sign Spotting Challenge focused on the problem of fine-grain sign spotting for continuous sign language recognition. We have released and made publicly available a new dataset of Spanish sign language of around 10 h of video data in the health domain performed by 7 deaf people and 3 interpreters. The added value of this dataset over existing ones is the frame-level precise annotation of 100 signs with their corresponding glosses and variants made by sign language experts. This paper summarizes the design and results of the challenge, which attracted 79 participants, contextualizing the problem and defining the dataset, protocols and baseline models, as well as discussing top-winning solutions and future directions on the topic.

TitelComputer Vision – ECCV 2022 Workshops, Proceedings
RedaktørerLeonid Karlinsky, Tomer Michaeli, Ko Nishino
Antal sider18
ForlagSpringer Science+Business Media
ISBN (Trykt)9783031250842
StatusUdgivet - 2023
Begivenhed17th European Conference on Computer Vision, ECCV 2022 - Tel Aviv, Israel
Varighed: 23 okt. 202227 okt. 2022


Konference17th European Conference on Computer Vision, ECCV 2022
ByTel Aviv
NavnLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Vol/bind13808 LNCS

Bibliografisk note

Publisher Copyright:
© 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.


Dyk ned i forskningsemnerne om 'ECCV 2022 Sign Spotting Challenge: Dataset, Design and Results'. Sammen danner de et unikt fingeraftryk.