[visionlist] ECCV’2022 Sign Spotting Challenge
Sergio Escalera
sergio.escalera.guerrero at gmail.com
Mon May 16 06:51:06 -04 2022
We cordially invite you to participate in our ECCV’2022 Sign Spotting
Challenge
Challenge description: To advance and motivate the research on Sign
Language Recognition (SLR), the challenge will use a partially annotated
continuous sign language dataset of more than 10 hours of video data in the
health domain and will address the challenging problem of fine-grain sign
spotting in continuous SLR. In this context, we want to put a spotlight on
the strengths and limitations of the existing approaches, and define the
future directions of the field. It will be divided in two competition
tracks:
1.
Multiple Shot Supervised Learning (MSSL) is a classical machine learning
Track where signs to be spotted are the same in training, validation and
test sets. The three sets will contain samples of signs cropped from the
continuous stream of Spanish sign language, meaning that all of them have
co-articulation influence. The training set contains the begin-end
timestamps annotated by a deaf person and a SL-interpreter with a
homogeneous criterion of multiple instances for each of the query signs.
Participants will need to spot those signs in a set of validation videos
with captured annotations. The signers in the test set can be the same or
different to the training and validation set. Signers are men, women, right
and left-handed.
1.
One Shot Learning and Weak Labels (OSLWL) is a realistic variation of a
one-shot learning problem adapted to the sign language specific problem,
where it is relatively easy to obtain a couple of examples of a sign, using
just a sign language dictionary, but it is much more difficult to find
co-articulated versions of that specific sign. When subtitles are
available, as in broadcast-based datasets, the typical approach consists of
using the text to predict a likely interval where the sign might be
performed. So in this track we simulate that case by providing a set of
queries (isolated signs) and a set of video intervals around each and every
co-articulated instance of the queries. Intervals with no instances of
queries are also provided as negative groundtruth. Participants will need
to spot the exact location of the sign instances in the provided video
intervals.
Challenge webpage: https://chalearnlap.cvc.uab.cat/challenge/49/description/
Tentative Schedule:
-
Start of the Challenge (development phase): April 20, 2022
-
Start of test phase: June 17, 2022
-
End of the Challenge: June 24, 2022
-
Release of final results: July 1st, 2022
Participants are invited to submit their contributions to the associated
ECCV’22 Workshop (https://chalearnlap.cvc.uab.cat/workshop/50/description/),
independently of their rank position.
ORGANIZATION and CONTACT
Sergio Escalera <sergio.escalera.guerrero at gmail.com>, Computer Vision
Center (CVC) and University of Barcelona, Spain
Jose L. Alba-Castro <jalba at gts.uvigo.es>, atlanTTic research center,
University of Vigo, Spain
Thomas B. Moeslund, Aalborg University, Aalborg, Denmark
Julio C. S. Jacques Junior, Computer Vision Center (CVC), Spain
Manuel Vázquez Enrı́quez, atlanTTic research center, University of Vigo,
Spain
--
*Dr. Sergio Escalera Guerrero*
Full Professor at Universitat de Barcelona
ELLIS Fellow / Head of Human Pose Recovery and Behavior Analysis group /
ICREA Academia / Project Manager at the Computer Vision Center
Email: sergio.escalera.guerrero at gmail.com / Webpage:
http://www.sergioescalera.com/ <http://www.maia.ub.es/~sergio/> / Phone:+34
934020853
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://visionscience.com/pipermail/visionlist_visionscience.com/attachments/20220516/0df15210/attachment-0001.html>
More information about the visionlist
mailing list