Wednesday, 28 Jun 2017
You are here: Home VCE Grand Challenge

Subscribe to our Newsletter

How many eyes has a typical person? (ex: 1)
Name:
Email:

Twitter Feed

Grand Challenge
VideoSense Grand Challenge 2014 PDF Print E-mail
Written by University of Reading   
Friday, 27 June 2014 07:55

VideoSense Grand Challenge 2014

Task Description    
Participant in this year Visual Privacy Task are invited to implement a combination of several privacy filters to protect various personal information regions in videos, by optimising the privacy filtering so as to: i) obscure such personal information effectively whilst, ii) keeping as much as possible of the ‘useful’ information that would enable a human viewer to interpret the obscured video frame.   Personal visual information is subjective human-perceived information that can expose a person’s identity to a human viewer.  This can include richly detailed image regions such as distinctive facial features or personal  jewellery as well as less rich uniform regions e.g. skin regions (that expose racial identity) or body silhouette showing a person’s gait (that generally helps to differentiate women from men and in some cases it may even enable a close friend or a spouse to identify the person).

In order to simulate context–aware privacy protection solutions, both Low or High information regions, and, unusual events e.g. fighting, stealing, and dropping a bag are annotated in the datset to be provided.   The participants are thus encouraged to exploit this information to achieve the appropriate level of privacy filtering for each person, object, and Low/High information regions  -to  optimise their privacy filtering mix according to criteria described under evaluation.

Registration
Task description: http://www.multimediaeval.org/mediaeval2014/visualprivacy2014/
Registration form: https://www.aanmelder.nl/mediaeval2014/subscribe

Target Group
Those working in  image/video processing and video-analytics for privacy protection applications.   

Task Schedule
6 May: Development data release
2 June: Test data release
11 August: Run submission due
15 September: Results returned
28 September: Working notes paper deadline

Task Organisers
Atta Badii (UoR), Touradj Ebrahimi (EPFL), Christian Fedorczak (Thales Communications & Security), Pavel Korshunov (EPFL), Tomas Piatrik (QMUL), Volker Eiselein (TUB), Ahmed Al-Obaidi (UoR).

Last Updated on Wednesday, 01 April 2015 09:35
 
VideoSense Grand Challenge 2013 PDF Print E-mail
Written by University of Reading   
Wednesday, 27 February 2013 09:48

This new VideoSense Grand Challenge builds on the results of the previous one (the 2012 MediaEval Privacy Task) and aims at finding ways to protect the privacy of people in videos. The insights arising from the Privacy Task as was run last year, showed that regardless of the methods used for filtering out the elements of the video image that can potentially reveal the identity of the data subject, the best results belonged to those using superior object detectors.  Hence this year the participants will be offered the object detections as given.

Accordingly, participants will be invited to propose methods whereby elements of the image of persons featured in video frames can be obscured so to render them unrecognisable. This is intended to ensure that a person appearing in a video frame will not be visually identifiable based on an image which is thus obscured.   This action is to be performed as a matter of privacy protection for any persons whose picture might happen to be knowingly or unknowingly captured in a video frame.  
The main elements to be obscured are as fallow: faces, ethnicity, gender, and accessories. You are free to filter other elements such distinctive clothing or gait, but should keep in mind that as much information as possible should be retained in order to keep the video understandable.

Since the resulting partly obscured videos would nonetheless have to remain available for viewing, an optimal balance should be struck so that despite the extent of such masking of the identity as may be necessary, the categorical identity of any masked data subject therein, i.e. as a human being, or the type of their activity as may have been evident in the original video frame, can still be recognisable to the viewer. Furthermore, the proposed obscuring techniques should take into consideration the acceptability and attractiveness of the resulting obscured and scrambled regions.

Participants are encouraged to present their results at the MediaEval 2013 workshop, which will take place in Barcelona, Catalunya, Spain, on Friday-Saturday 18-19 October 2013. Note that this is just before ACM Multimedia 2013 http://acmmm13.org/, which will be held Monday-Friday 21-25 October 2013, also in Barcelona.

Provisional Task schedule
June 3rd: data set release
September 13th: run submission
September 25th: results returned
October 18th: MediaEval workshop

Registration will soon be open for participants at http://www.multimediaeval.org/mediaeval2013/

Last Updated on Thursday, 28 February 2013 09:19
 
VideoSense Grand Challenge PDF Print E-mail
Written by University of Reading   
Thursday, 02 February 2012 10:14

VideoSense Grand Challenge now open !    

The VideoSense European Network of Excellence has a mission to promote research collaboration focused on solution systems (algorithms and architectures) that support socio-ethical video-analytics. As part of such activities this year we have included within the MediaEval 2012 a relevant Challenge for the research community by way of a problem-based task referred to as the “Visual Privacy” Task seeking solutions for Situated Privacy-aware Video-Analytics.  We hope you will find this challenge interesting and inspiring and we very much like to encourage you or a member of your research team to consider participating.   For this you will need to visit the MediaEval 2012 website and register for the “Visual Privacy” Task.

Task description: http://www.multimediaeval.org/mediaeval2012/visualprivacy2012/index.html

Registration Form: http://www.aanmelder.nl/mediaeval2012/

The registration is open till the end of April 2012.

 

 


The VideoSense European Network of Excellence has a mission to promote research collaboration focused on solution systems (algorithms and architectures) that support socio-ethical video-analytics. As part of such activities this year we are including within the MediaEval 2012 a relevant Challenge for the research community by way of a problem-based task referred to as the “visual privacy task” seeking solutions for Situated Privacy-aware Video-Analytics.  We hope you will find this challenge interesting and inspiring and we very much like to encourage you or a member of your research team to consider participating.   For this you will need to visit the MediaEval 2012 website and fill in the survey that is to collect your opinion on which tasks should be run in MediaEval 2012:
 
https://www.surveymonkey.com/s/mediaeval2012tasksurvey
 
The survey will take you 5 minutes if you fill in only the main questions and about 25 minutes if you chose to answer the additional questions and give comments (which we encourage).
 
The MediaEval 2012 task list will be finalised by mid-February and the sign-up phase will open soon after.
 
Please be sure to fill your email address in on the first page of the survey if you would like to receive an mail when the sign-up opens.
 
See http://www.multimediaeval.org for further information. Note that the run for this year will be due early to mid-September and the workshop will be held early October (i.e., a month later than last year).

Last Updated on Wednesday, 27 February 2013 09:47