====================================
CALL FOR TASK PARTICIPATION: MART (Micro Activity Retrieval Task) at
NTCIR-15
====================================
*Registration deadline:* 15 July 2020 (extended)
*Paper due:* 20 Sep 2020
*Camera ready due:* 1 Nov 2020
*NTCIR-15 Conference:* 8-11 Dec 2020 (NII, Tokyo, Japan) (remote online
presentation option available)
*MART* (Micro Activity Retrieval Task) is an NTCIR-15 data challenge task.
MART aims to motivate the development of a first generation of techniques
for high-precision micro-activity retrieval, that is activities that occur
over short time-scales such as
minutes. Participating researchers will develop and benchmark approaches to
retrieve micro-activities from *rich time-aligned multi-modal sensor data
(provided)*. The dataset to be used for MART has been collected by
equipping individuals to gather a detailed lifelog as they progress through
a protocol of real-world activities e.g. using a computer, solving a
problem, eating, daydreaming, etc. The labelled training dataset includes a
lifelog camera data stream, (bio-)signal activity (EOG, HR, accelerometers,
etc), and computer access to record interactions (mouse movements,
screenshots). This task will present a new challenging set of
micro-activity topics.
*There are two sub-tasks in MART @ NTCIR-15:*
A) *Micro Activity Retrieval Task*: where participating groups'
approaches will be benchmarked via mAP.
This task may be undertaken in an interactive or automated manner. A
baseline interactive system is provided on which task participants can
build upon.
B) *Micro Activity Insights Task*: where participating groups will be
encouraged to find interesting aspects of the task dataset
that will provide insights for researchers in how multi-modal sensing
(and approaches) can support future (micro-)activity detection/retrieval
tasks. *Contributions exploring such aspects of the dataset are highly
encouraged.*
Participating organisations will be expected to submit a paper outlining
their approach that they will then present at the MART workshop in Tokyo,
Japan this coming December 2020. *Please Note:* Due to uncertainty
regarding international travel and COVID-19, MART @ NTCIR-15 will have a
remote (online) presentation option for task participants.
We provide a working baseline approach to enable participating researchers
to build upon an already working approach (along with a training dataset).
For more information please visit: http://ntcir-mart.computing.dcu.ie/
To get access to the data/participate in the evaluation task, please
complete the signup procedure via the NTCIR-15 website and contact the task
organiser Graham Healy ([log in to unmask]) for dataset access:
http://research.nii.ac.jp/ntcir/ntcir-15/howto.html
Important Dates:
15 July 2020 - Task Registration Due
1 Aug to 31 Aug 2020 - Formal Run
31 Aug 2020 - Evaluation Results Release
20 Sep 2020 - Paper Due
1 Nov 2020- Camera ready due
8-11 Dec 2020 - NTCIR-15 in NII, Tokyo
Regards,
Graham Healy
--
*Graham Healy*
Assistant Professor | School of Computing | Dublin City University
http://www.computing.dcu.ie/~ghealy
[log in to unmask] | 01 700 7364
School of computing | L2.36 | McNulty Building | Dublin City University |
Glasnevin | Dublin 9
--
*
*Séanadh Ríomhphoist/Email Disclaimer*
*Tá an ríomhphost seo agus aon
chomhad a sheoltar leis faoi rún agus is lena úsáid ag an seolaí agus sin
amháin é. Is féidir tuilleadh a léamh anseo.
<https://sites.google.com/view/seanadh-riomhphoist>*
*This e-mail and any
files transmitted with it are confidential and are intended solely for use
by the addressee. Read more here.
<https://sites.google.com/view/dcu-email-disclaimer>*
*
--
<https://www.facebook.com/DCU/> <https://twitter.com/DublinCityUni>
<https://www.linkedin.com/company/dublin-city-university>
<https://www.instagram.com/dublincityuniversity/?hl=en>
<https://www.youtube.com/user/DublinCityUniversity>
---------------------------------------------------------------
For news of CHI books, courses & software, join CHI-RESOURCES
mailto: [log in to unmask]
To unsubscribe from CHI-ANNOUNCEMENTS send an email to
mailto:[log in to unmask]
For further details of CHI lists see http://listserv.acm.org
---------------------------------------------------------------
|